What is abstract labour, and who does the abstracting?

Money: the form of value

We all know that parts of reality represent or measure other parts of reality. A ruler measures length, a thermometer measures temperature, and so on. We created these  measuring devices for a definite purpose.

But the meaning of money, what it might signify or represent, is less clear. Money is clearly a human technology that first entered the historical stage over 2000 years ago. But precisely why humans “create” money, and what its symbol may represent, remains a subject of controversy.

To be clear, by “money” I don’t mean actual coins or notes but instead the numerical quantities we see stamped on coins or printed on notes, or stored as bits in computers, and so on. To be precise, I should really say “unit of account”. But saying “money” is simpler, as long as we’re clear about what we mean.

Now, Marx tackles the meaning of money in his famously difficult, opening chapters of the first volume of Capital. He notes that the exchange of commodities in the market implies there’s something equal, or equivalent, about them. For example, if I sell 20 yards of linen for 10 pounds, and then spend my 10 pounds on a new coat, then, indirectly, 20 yards of linen have been made equal to 1 coat by the act of exchange.

If market prices were entirely random there would be nothing to say because this equivalence would be accidental. But although prices fluctuate they are not random. There is a strong signal in the noise. Typically, you can’t sell a pen and then buy a plane. And you can’t work for a day and then spend your day’s wag to buy a mansion. There are exceptions. But the exceptions prove the rule.

So during any period of time there are definite well-established market prices that determine the ratios in which commodities can exchange, that is are equalised, with each other.

The magical affinity of all things

A quick dip into any anthropological textbook quickly reveals that humans entertain the most diverse and extraordinary beliefs about how the world works and how we should conduct our everyday lives. What some cultures consider normal, others would consider strange and bizarre.

We rarely take an anthropological viewpoint on our own culture. That’s because it’s hard to do. It requires stepping out of one’s conceptual framework, and looking at the ordinary and accepted as unusual and questionable. And we’re not educated to do that. In fact, we’re generally educated to do the opposite.

Only fervent occultists, keepers of highly esoteric knowledge, would dare claim that everything we see around us, all the things and activities in the world, are — despite all appearances — really the same. That 1 kg of caviar, fished from the Caspian Sea, is “the same as” 1000 different people, living in different locations around the world, clicking on the same internet advert. Or clowning at a children’s party is actually “the same as” 200 rounds of shotgun ammunition. Or that 1 month of computing time on a high-spec machine in the cloud is “the same as” 1 tonne of potatoes. Only the highest adepts, schooled in years of training, devotion and meditation, could see the truth of such magical affinities and identities.

But we more than see the truth of it. We openly and regularly achieve it. We manifest these magical affinities between diverse things on a daily basis. We treat quantities of fish eggs, human attention, clowning performances, bullets, computing time, potatoes, and a bewildering array of other things, as “the same” — because, in the marketplace, they all may be exchanged for one another, mediated by this mysterious technology we call money.

Traditional magic rather meekly only proposes the existence of affinities between planets, minerals and human fate. But the magical operations of our modern commercial world — where every thing, activity and even future event is successfully reduced to comparable quantities of this substance we call “money” — overwhelmingly surpass, in both scale and ambition, the most deranged fantasies of the medieval grimoires. Market exchange achieves a universal affinity between all things under the sun.

The economic mysteries

It is for these sorts of reasons that Marx writes of the “mystery of commodities” with its “magic and necromancy“.

Market societies achieve a titanic conceptual abstraction: every single thing that we swap between ourselves is stamped with a single quantitative property that we call exchange-value. But, rather mysteriously, no single person, no single consciousness, is responsible for the abstraction that we call exchange-value. And, in the vast majority of cases, no human institution has the goal, or the power, to set prices. No-one is controlling this abstraction.

So we have two economic mysteries: a ubiquitous social abstraction without any obvious content, and an abstraction without an abstractor.

So I want to dispel some of the mystery, but also add to it. Scientific work doesn’t only explain, but also discovers things that we didn’t notice before. In this talk I will to try to answer two questions: what does the abstraction of exchange-value represent? and who, or what, is doing the abstracting?

The content of value, or abstract labour

So let’s begin with the first mystery. What is this abstraction? What do those money quantities actually denote?

Marx argues that exchange-value refers to a special, common property shared by all commodities — that of being the products of labour. So caviar and clicks are the same because, to manifest them as commodities in the marketplace, requires the sacrifice of someone’s labour.

Now it must be admitted that Marx’s argument — for the proposition that the special common property shared by all commodities is labour — is unsatisfactory. But fully stepping into the whys and wherefores of Marx’s argument would, I think, divert us. So I’ll simply state my opinion, which is this: Marx’s conclusion is correct, but his argument for it isn’t. And given that, let’s just accept Marx’s conclusion for now, as a working hypothesis, and keep going.

Marx then says that the common property cannot be specific kinds of labour — because fishing for caviar, or writing advertising software, or clowning, or making bullets — are very different activities.Β 

The act of exchange abstracts from the individual peculiarities of different labouring activities, leaving something common to all of them, which Marx calls “human labour in the abstract”, or abstract labour. Commodities, according to Marx, have economic value “only because human labour in the abstract has been embodied or materialised in it”.

Now, we have to be careful with the term “embodied”. Marx doesn’t literally mean that abstract labour inheres within the material body of the commodity. Abstract labour is not a physical property of a thing. What he means is that some fraction of the total labour time of society must be used-up, or expended, to produce the commodity and bring it to market.

So abstract labour is not concrete labour, not a specific type of labouring activity, but something else, something deeper and more general. As Marx states, abstract labour has “the character of the average labour-power of society”. So a good first approximation is for us to think of abstract labour as denoting the casual powers of the typical or average worker. That isn’t quite right, but it will do for now.

So, according to Marx, the titanic abstraction achieved by commodity exchange refers to a specific content, which is a property of the material world that he calls abstract labour.

How do we measure abstract labour?

Marx, in the opening chapters of Capital, then immediately asks the obvious question, “How, then, is the magnitude of this value to be measured?” and he answers, in a seemingly straightforwardly way, that it is measured “by its duration, and labour time in its turn finds its standard in weeks, days, and hours.” So we’re talking about units of time.

We might suppose, therefore, that we can immediately pull out our stopwatches and start measuring the amount of time people spend working, and then correlate our measurements with the prices we observe in the market. Because if prices really do represent labour-time then we should, in-principle, be able to scientifically verify this claim.

But that would be too hasty. Before we can even consider empirically verifying Marx’s theory of value, we need more clarity on what that theory actually is. Theory has to precede measurement.

Now I’m not sure how deliberate this is, especially as I read Marx in translation. But it might be noteworthy that Marx does not ask, “How should we measure quantities of abstract labour?”, and neither does he answer by saying that “we can measure it by its duration”.

And that’s because we don’t measure abstract labour. Something else measures it.

This property of Marx’s theory — that money refers to labour time in virtue of our collective, social activity and independently of our thoughts about it — is radically different from the classical political economy of his day, and also modern economic theory.

The abstraction is not ours because our cognition is not performing the abstraction. We are not the abstractor. Instead, the mysterious abstractor is taking the measurements about labour time and connecting the form of value, which is money, to its content, which is abstract labour.

So, as scientists, our first job isn’t to start measuring labour time. Our first job is to understand what the abstractor is, and how it connects its abstraction to its world. We need a theory of this entity, and its powers, before embarking on empirical verification.

Who or what is the abstractor?

So we have a partial answer to the first economic mystery. The abstraction of exchange-value, or more plainly money, refers to “abstract labour”. So let’s turn to the second mystery: who is doing the abstracting? who or what is the mysterious abstractor?

In fact, Marx has already told us who it is. Sometimes mysteries hide in plain sight. The big clue is Marx’s choice of the title for his magnum opus. The abstractor is what Marx calls “capital”.Β 

But the term “capital”, especially now, doesn’t evoke the right images and associations. First of all, it gets us thinking about large sums of money. A capital sum. But Marx isn’t really talking about big sums of money. And, second, modern economic theory has reduced the term “capital” to a anodyne accounting term that manages to mix-up, in a confused way, capital equipment with large sums of money.

But capital, for Marx, is first and foremost a social practice. Capital denotes a collection of activities that certain people regularly do embedded within a system of property rights, contracts, and coercive power. Capital is a circuit, where an initial capital sum is “invested” in production, and then typically returns with a profit increment. Capital enlarges itself, whenever it can. This circuit is mediated not only by money, but also economic production itself, including the disciplining and exploitation of workers.

Marx’s language — of capital, of social relations of production, circuits of accumulation, and so on — doesn’t, in my opinion, fully evoke what’s going on. So instead of saying “capital” I’m also going to say “the controller”. Because capital is a control system, not merely in the political sense, but in the more profound and scientifically important sense of being a negative feedback control system. Capital is literally a controller. So if capital is a controller, then how does it work, and what does it control?

Capital is a negative feedback control system

In a fully capitalist system, profit-incomeΒ must be reinvested in order to make more profit. This is the prime directive for anyone who possesses a capital sum of money. If this directive isn’t followed the capital will rapidly diminish and expire.

Owners of capital β€” that is capitalists β€” can’t put all their eggs in one basket. That’s too risky because firms can go under, and assets might depreciate. So capitalists own a portfolio of investments with different risk profiles, such as government bonds, shares in different companies, and more speculative bets in high-growth sectors.

Each individual capital attempts to maximise the return over its portfolio. If it fails it will diminish, and eventually cease being a capital at all.

And it’s right here that we find the casual structure of a feedback control system. Capital has its own goal state, sensory inputs, decision making, and ability to act upon the world in which it is embedded.

Let’s take each of these in turn. (i) The goal of an individual capital is to maximise the average return from every dollar (or pound) invested. (ii) The β€œsensory inputs” are the different profit-rates earned across the portfolio. (iii) The capitalist, or the financial experts they employ, compare the different profit-rates, and (iv) the feedback loop is closed by actions that withdraw capital from poorly performing investments, and inject capital into high performing investments.

This control loop manifests as an insatiable and ceaseless search for high returns.

The control loop doesn’t care how its capital is actually used in production. It entirely abstracts from all concrete activities. The only thing it can sense, compare and use is abstract value.Β 

So the commanding heights of the global economy consists of an enormous ensemble of individual capitals, each manically scrambling for profit, reacting to theΒ signals of differential returns received from its tendrils that extend to every productive activity under its rule,Β continually injecting and withdrawing capital to and from different industrial sectors and geographical regions. The entirety of the world’s material resources, including the working time of billions of people, are repeatedly marshalled and re-marshalled away from low and towards high-profit activities. In the space of months, entire industrial sectors may be raised up, relocated, or thrown down.

Capital tends to concentrate in few hands. In consequence, the scale and power of some capitals rivals the Titans of old.

All these autonomous control loops have the single-minded goal of extracting profit from the world’s activities. If an activity fails to satisfy this goal, then the controller withdraws its capital, and the activity stops.

So at the apex of the economy we have a competing collection of identical controllers β€” with an atavistic, low level of intelligence β€” which inject and withdraw a social substance that appears to possess the magical power of animation, of bringing things alive, of creation; but also appears to possess the demonic power of annihilation, of suffocation, of bringing things to an end, of destruction.

The controller and the controlled

So capital is a control system. The money substance, which it deals in, flows into and through every aspect of our social intercourse, like the blood in veins. But what does the controller really control?

Sometimes it’s obvious what a particular control system controls, because we designed it. For example, we know that a thermostat controls room temperature, and that a governor controls the speed of a steam engine.

But the vast majority of control systems are not designed by people. Nature is stuffed full of them, from simple homeostatic mechanisms to incredibly complex animal brains. These systems evolved, without a designer, and therefore our we must work harder to determine what they control, and what their internal representations may, or may not, represent in the environment in which they act.

I will skip the details of a scientific theory to determine what controllers in fact control. It’s not a simple story. I think the complexity of that story partially explains why Marx’s argument that abstract labour is the substance, or content, of value, in the opening chapters of Capital — chapters that he famously worked and re-worked, that Engels joked bore the marks of Marx’s painful carbuncles — is not entirely satisfactory. Marx had stumbled upon a hard problem that couldn’t be fully solved with the conceptual tools of his time.

So, I’ll skip the methodological detour, and instead jump to the conclusion and simply state what capital, as a control system, in fact controls.

We already know that capitals, both big and small, are intimately connected to the process of production. The capitalist firm needs to borrow capital to buy inputs and means of production, and hire workers. Workers supply concrete labour to produce specific use-values for sale in the market.

Now, the controller judges all the different concrete activities occurring across its portfolio in the same way: which activities yield above-average returns, and which do not? The controller rewards firms that make comparatively high profits with new injections of investment; but punishes firms that make comparatively low profits, or losses, by withdrawing its capital. These monetary rewards and punishments flow down, through firms, into the labour market, and reward concrete labour by the payment of wages, or punish by withdrawal and unemployment.

In this very real sense, capitalΒ wants specific kinds of concrete activities, and does notΒ want other kinds. The kinds of activities it wants are those that yield above-average profit. Capital is therefore controlling us. It controls how we spend our time.

Abstract labour: the kind of labour that capital wants

So what kinds of activities does capital want? Simplifying somewhat, we can identify two essential properties that concrete labour must possess in order to yield profit.

First, the concrete labour must be useful to others, that is produce commodities that can be sold in the market. No-one will buy a coat with three arms.

Second, the concrete labour must have above-average efficiency; in other words, a firm makes more profit if it uses-up less labour-time than competitors that produce the same commodity.

And this is why, just after Marx first introduces the concept of abstract labour, he immediately points out that only socially necessary, and useful, labour counts as abstract labour.Β 

Capital does not want workers to spend time smelling the roses with their family and friends. That activity doesn’t yield saleable use-values. Neither does capital want workers to slack on the job, or become ill. Slacking or illness isn’t efficient. Capital, if it completely had its way with us, would have us spend all our time labouring in the firm at the highest possible intensity, continually striving to out-compete other workers in the labour market. This is the kind of behaviour that capital wants.

So capital controls concrete labour, the real labouring activities of the working population in all their diverse manifestations. And capital controls actual labour-time, actual clock times of real people doing real things. It is capital itself that holds a metaphorical stopwatch in its hand, measuring and accounting, and judging and condemning; always on the look-out for the slightest slacking off or insubordination.

And the goal of capital is to convert concrete labour into abstract labour, into the kind of labour that both fits into the division of labour, so it can be exchanged against other labour, and into the kind of labour that fully sacrifices itself to capital, gives itself up as tribute, in order to yield profits for the capitalist firm, and ultimately the controlling, dominant capitals.Β 

In other words, “abstract labour” is manifested, brought into reality, by capital itself. Maximising profit is identically the process of maximising the manifestation of abstract labour out of concrete labour.

This is why Marx says that only abstract labour “creates value”. Concrete labour may or may not create value. If it doesn’t, it isn’t abstract labour, and capital as controller quickly works to eradicate its existence, by withdrawing capital from the firms that employ it.

An alien cognition that binds form and content

I think that viewing capital as a negative feedback control system is the best way to understand what it actually is, and the fundamental role it plays in shaping our society and the kinds of lives we lead. The language of control systems is really just a more precise way of talking about the dialectical relationship between the form of value and its content.

So capital is a controller that employs a form of value — money — to control the content of value — which is our labour time. The form and the content are bound together, linked semantically in a relation of representation to referent, by the lawful regularities instantiated by generalised commodity production.

Control systems instantiate atavistic versions of the basic elements of cognition. They in fact sense, decide, and act. They in fact have internal representations that refer to the world they act in. In consequence, Marx’s theory of value is fundamentally a theory of an alien cognition that control us. His instincts were “on the money” when he wrote of the necromancy of commodity production because only the religious, magical and occult traditions in our history have adequate concepts to express our predicament.

Capital is an egregore that controls the manifestation of abstract labour from concrete labour. It rewards concrete activities that yield the tribute of surplus-value, and punishes those that don’t. Painting by Johfra Bosschart.

An egregore is an non-physical entity that exists in virtue of the collective ritual activities of a group yet operates autonomously, according to its own internal logic, to materially influences and control the group’s activities. The group creates the egregore, and the egregore creates the group, in a self-reinforcing feedback loop.

In traditional cults the egregore typically takes the form of a god. However, the ritual activities of the initial capitalist cults were so materially successful they rapidly metastasised and, in a few centuries, engulfed the world. What is universal becomes the unnoticed background. So the egregore, in our society, is hard to see. It hides in plain sight. We refer to it, of course, but obliquely, using soporific names, such as “the economy” or “capital”. We need a better name for it, one designed to wake us from our slumber.

Capital is an egregore. Not metaphorically, or ironically, but actually. Capital is a being, an autonomous entity, with primitive thoughts about us. Money is how it measures us, and money is how it commands us. Capital is an alien cognition that acts in the world to bind the form of value to its content.

Neither naive materialism nor subjective idealism

So we know what the abstractor is now. And now that we have a clearer grasp of the core structure of Marx’s theory of value it becomes much easier to spot misinterpretations of it.

There are misinterpretations that emphasise the content at the expense of the form. Marx’s theory is not at all like the naive materialism we find in classical political economy, or modern Sraffian interpretations of Marx, which posit one-way causation from labour-time to money prices. Instead, we must think about feedback loops, about two-way causation, from content to form, and from form back to content.

But there are other misinterpretations that emphasise the form at the expense of the content.

Clearly, Marx’s theory is an objective theory of value. Despite the pretensions of subjective utility theories of value we cannot collectively wish planes to be cheaper than pens. We are not the dominant controller, we are the controlled. The individual consumer is not king.

But more sophisticated variants of subjectivism also misinterpret Marx’s theory. Some Marxists think capital dreams about abstract labour, that abstract labour is an invention of the capitalist system, which doesn’t actually refer to something existing independently in objective reality. This is a kind of social constructivism, where Marx’s theory is reduced to a postmodernism parody of ghostly and ideal forms.

In these misinterpretations the form has no content. And so money doesn’t refer to any property that exists independently of it. The form creates an illusory content. In this view, abstract labour may indeed have real effects, in the way that belief in an Father Christmas may cause people to offer cookies and milk, but it doesn’t really exist. This seems like a sophisticated position but ultimately reduces to value nihilism, where there are only prices, and there is nothing hidden behind them.

But Marx’s theory is essentially about the control of concrete labour time, the actual objective working conditions of millions of people. Any interpretation of Marx that claims abstract labour cannot be measured independently of markets and prices, or cannot provide a definition of the content of value without relying on magic coefficients that depend on prices — has gone awry. (Perhaps not too surprisingly, both naive and sophisticated misinterpretations of Marx are rife in the bourgeois academy).Β 

Of course, like any entity, capital’s thoughts about may not perfectly reflect, or represent, the reality in which its embedded. However, if a control system successfully controls then its internal representations will bear a truthful correspondence to reality. And capital is a supremely successful controller.

And, ultimately, this is why we can verify Marx’s theory: labour is already disciplined to be efficient and useful. And so the majority of concrete labour is already abstract labour. In consequence, if we pick a group of 50 workers randomly they will approximate the value-producing power of 50 units of abstract labour. Take larger aggregates and the approximation only improves. Taking out our stopwatch won’t work at the level of an individual worker because there’s no guarantee their concrete labour will ultimately count as abstract labour. But our stopwatch will measure abstract labour if we collect sufficient samples. As Marx stated, abstract labour has the character of the average labour-power in society. So the control success of capitalism means we can measure quantities of abstract labour before that labour is equalised and homogenised in the market.

An analogy might help here. An ethologist, studying the behaviour of an animal in the wild, can’t truly get inside the animal’s head and see the world through its eyes. The ethologist can never fully know what it’s like to be a bat. But nonetheless ethologists have developed detailed theories of echolocation, and how a bat’s cognition represents its environment. In a similar way, we are studying the behaviour of an autonomous entity, called capital, with an alien cognition. Abstract labour is its concept, not ours. But we can form a concept of abstract labour that corresponds to its concept of abstract labour. After all, we, the controlled, and it, the controller, all live in the same world. And we can both talk about, and represent, an objective property of it.

And what is that objective property? We can now refine our initial, approximate definition of abstract labour. It is not just average labour, or the common causal powers of human labour. It is something more specific, something more historically determined and contingent.

Abstract labour is a collection of causal powers possessed by human labour that can manifest as an ability to produce an endless variety of useful things for others, to make profits by working harder or longer, to improve techniques of production so more may be produced with less, and to out-compete others in a ceaseless scramble for profit. If we workers lacked these causal powers then capital would fail to mould us into the value-creating, homogeneous units that it wants. (And, as an aside, it is precisely because machines, and animals, and other non-human factors of production, lack these causal powers, lack our divine spark, that their “labour” cannot, and does not, count as abstract labour.)

Children of the demiurge

OK, let’s draw to a close, and try to sum up.

We asked, at the beginning, what money represents, and the answer is abstract labour, which is labour that creates profit for capital. And we asked who does the abstracting? And the answer is capital does that, when we understand capital to be an autonomous controller that supervenes upon our social practices, but is not reducible to them.

The value form, the titanic abstraction that controls our lives is, in a sense, the primitive language of the controller. It sees and judges our activities in terms of abstract value, by comparing differential profit-rates across its portfolio. But it also commands our activities using abstract value, by injecting and withdrawing its substantial being, which is money. Capital commands us. Our real, living, concrete labour is the object of its control: it works to mould, shape, and discipline the total labour-power of society into the specific form of abstract labour, which is labour that gives itself up, utterly and completely, as tribute to capital.

So the value form participates in both measuring labour time, and also commanding labour time. We shouldn’t be surprised that the value form also has imperative semantics. Generalised commodity exchange has no conscious planner or plan, and therefore the command and control necessary to organise the division of labour is achieved through the allocation of capital, the transmission of money and the structure of prices.

Capital commands concrete labour time to manifest as abstract labour time, and therefore brings into being what is already latent within us. But capital intensifies and perfects only a part of us. We are more than merely creatures able to manifest abstract labour. We have the power to do much more than merely produce useful things by working intensely for long hours. Of course, despite capital’s rule we carve out pockets of resistance, where we may be more fully ourselves. But capital does not want us to play, learn, explore, care or give freely. Capital wants us to produce, endlessly. And therefore we, under the rule of capital, are reduced to shadows, mere narrow abstractions, of what we could be.

Allow me to finish with a very blunt analogy. Cows can do lots of things. But all we care is that they produce as much milk and meat as possible. And so we breed them, inject them, rear them, and control them to do only that. Sometimes their udders are so distended by excessive production they tear, split and spill.

We are cattle to capital. We too have become distorted and disfigured by its rule. It brands us abstract labour. But we are also concrete individuals. The form does not exhaust the content. And this seemingly innocuous non-identity between form and content is the fundamental reason why, one day, we will escape from capital’s rule.


This is Part 3 of a connected series (that can be read in any order):

Part 1: What is the meaning of money?

Part 2: Prolegomena to a demonology of capitalism

Further reading

Wright, 2014.Β Loop-closing semantics. In: From animals to robots and back: reflections on the hard problems in the study of cognition. Cognitive Systems Monographs: Springer, pp. 219–253.

Rubin’s “Essays on Marx’s Theory of Value”

Karl Marx’s invisible hand

Wright, I. (2017) The general theory of labour value.

Wright, I. (2015) The Law of Value: a contribution to the classical approach to economic analysis

Pilling, G., 1986. The law of value in Ricardo and Marx. In: Fine, B. (Ed.), The Value Dimension – Marx versus Ricardo and Sraffa. Routledge and Kegan Paul, London and New York, pp. 18–44.

Evocation of the demiurge in the magic circle of production, exchange and consumption. Members of our Society observe this ritual on a daily basis. The demon manifests inside the circle and reveals the secret affinity of all things. The great beast takes many forms, commands uncountable legions, and wields the power to transmute lead (concrete labour time) into gold (abstract labour time). Initiate, if your sacrificial offering is judged worthy, you may receive a small share of the gold, otherwise expect nothing. A small number of adepts invoke the demon and join the ranks of his possessed legions. Thereupon they boast of receiving the greatest riches, yet at great cost to their souls.




  1. Hi Ian,

    Really nice post once again! I’m waiting to see where your “dark marxism” series is heading. I wish you’ll find a way out from the ecoteric world of capitals and abstract labour eventually.

    Also, I think your formulation that “Marx’s theory of value is fundamentally a theory of an alien cognition that control us” is spot on. It really is the starting point for proper understanding of ideas about “real abstraction” and that philosophical Marxist stuff.

    However, there were some formulations that didn’t sound quite right to me in your talk. Just three minor comments.

    1) “For example, if I sell 20 yards of linen for 10 pounds, and then spend my 10 pounds on a new coat, then, indirectly, 20 yards of linen have been made equal to 1 coat _by the act of exchange_.”

    But Marx’s point was the opposite. 20 yards of linen and one coat are not made equal by the act of exchange, but they, rather, appear to be made equal by it. What this appearance, ie., equality by the act of exchange, manifests is a deeper equality at the point of production between different abstract labours.

    Rubin never grasped this point I think, but to me it seems to the most essential point, for it warrants that it is competition that is the mechanism of equalisation of labours, not exchange that merely mediates competitive process.

    2) “Each individual capital attempts to maximise the return over its portfolio. If it fails it will diminish, and eventually cease being a capital at all.”

    Surely, you should have written “each individual capitalist”, right? Individual capitals are combinations of use-values (industrial plants in large-scale production), and their operation conditions determine real costs for the production of commodities. As these costs differ, but prices are forced towards equality by competition (within industries), profit rates of individual capitals tend to differ systematically.

    So, it is capitalists, or capitalist firms, that shepherd these collections of capitals, and breeding them like farmers breed stock of cattle. The idea of portfolio investment, which reflects contemporary financialized forms of capitalism, seems like a secondary point here, the crucial point being real competition between capitals allowing capitalists to spot differences (benchmark profit rate minus individual profit rate) within differences (price minus costs).

    In this sense individual capitals are completely sightless, lacking any cognition and expanding blindly, but as capitalists and managers of capitalist firms (also financial firms) use their cognition in social practices of profit seeking (however imperfectly), they let themselves to become vechiles that lead to the emergence of the alien cognition of the capitalist system or capital in general.

    But here, again, it is real competition that seems to be the organising principle.

    3) “(And, as an aside, it is precisely because machines, and animals, and other non-human factors of production, lack these causal powers, lack our divine spark, that their β€œlabour” cannot, and does not, count as abstract labour.)”

    This seems like a hopeless way to go in my humble opinion. The reference to “divine spark” is just mysticism, not mysticism created by capitalism and trancended by dialectical analysis, but mysticism pure and simple.

    I think Marx’s point was that labour time is the measure of value, because capitalist commodity production system is essentially a social division of labour, and law of value emerges from the working of this peculiar division of labour as you yourself have shown.

    If humans were replaced in production by autonomous, self-learning and self-replicating machines (like Homer’s self-moving tripods mentioned by Aristotle), the cost of reproduction of these machines, and “the necessary robot time” of that it took from these robots to manufacture particular products, would still materially constrain production process, thus, being a kind of “measure of value”. But labour time of humans would become redundant. And this, it appears to me, would also make capitalist relations of production redundant.

    But what social arrangement would come into its place would remain an open question. And I think it is quite possible that such technique could become an alien power even greater extent that capitalism as it stands today.

    Liked by 1 person

    1. Hi Miika

      Thanks for your insightful comments!

      Yes, I think the “control system” point-of-view on Marx’s value theory is clarifying. I should have mentioned “real abstraction” in my post.


      I’m unsure what to make of your point (1). Commodities of equal exchange-value not only “appear” to be equal but are in fact equal (in terms of exchange-value). I do agree, however, that this equality in the market points to a “deeper equality at the point of production between different abstract labours”. Yes, absolutely.

      I think you critique of Rubin is interesting, but I don’t yet fully understand it. Rubin, precisely because he had a deep understanding of classical political economy that informs Marx’s work, is brilliant on the law of value as a mechanism for the allocation of the division of labour within society via the “transmission belt” of money and exchange-value. But, yes, he doesn’t spend as much time on the competition between capitals, and the drive to intensify exploitation. I agree that exchange, in itself, doesn’t equalise labour. But Rubin’s eplanation of the law of value does emphasise (as he must) that it’s fundamentally a competitive mechanism.

      Is there any literature you can point me to that expands on your point (1), particularly the concept of “real competition”?


      In my terminology “capital” denotes a social practice with multiple moments that take different forms (capital sum, use-values, labour process, etc.) as per Marx’s lengthy discussions of the circuit of capital (and the myriad forms of value) in Volume 2. I am not using “capital” to denote capital equipment, as per standard economic theory.

      I agree that individual capitalists “shepherd” the capital, and “use their cognition” to seek profits. But we cannot reduce social practices to individual consciousness. The social practice of “capital” (the M-C-M’ circuit) instantiates a control system that’s not reducible to the individual cognition of the humans that it supervenes upon. This control system instantiates primitive forms of cognition that, again, are not reducible to human cognition. Capital, in this egregoric or animistic sense, is not “sightless” but on the contrary really does have primitive analogues of sensing, thinking, and acting. In fact, capital itself has automated some of the human cognition that it was once implemented upon (e.g., algorithms that automatically allocate capital have partially replaced fund managers).


      I see that my reference to a “divine spark” didn’t work for you. Just to be clear: the only possible materialist explanation why labour is the source of value, and not any other factor of production, must be in terms of the causal powers of human labour and their relation to the controller. So I completely agree when you say, “if humans were replaced in production by autonomous, self-learning and self-replicating machines” (i.e., machines with casual powers identical, or superior, to humans) then the labour of these kinds of machines would also constitute value (i.e. function as abstract labour). However, no machine today comes close to the causal powers of abstract labour (and I say this as someone who has worked in machine learning their entire professional career).

      This issue — of how and why only humans are the source of value — deserves an entire post on its own. So we’ll just have to be brief for now. If I adopt, as you do, an entirely materialist perspective on this issue then why did I use the phrase, “divine spark”?

      I’m deliberately referring to the Gnostic idea that a portion of the godhead resides in every human being. Obviously, I do not believe in the existence of a divine creator from which we’ve inherited some of its properties. That would indeed be a religious or mystical claim (because no experiment could be designed to test the proposition). Rather, I wish to obliquely point to an analogous claim: that the casual powers of human labour are “universal” in the sense of computability theory: that in principle we have the power to understand any aspect of the material world, and control it. This is closer to the traditional dialectical claim that thought and being are, ultimately, identical. So our cognition and our powers are as divine as the entire cosmos itself. The form does not exhaust the content. Contrast this point-of-view to how labour is metaphorically portrayed in bourgeois commercial culture, where labour is just another, entirely profane, “factor of production”, on a par with land and machinery, and all creativity and innovation is either ignored or shifted to a heroic entrepreneur (endowed with capital).

      Hope these comments are helpful, and thanks for engaging.

      Best wishes,


  2. Hello prof.wright reading your post and watching the audio of your other post it seems clear to me that you are drawing from the ”total capital” , ”capital in general” and ”individual capitals”, ”capital in compettition” distinction that a lot of marxists subscribe to https://link.springer.com/chapter/10.1007/978-3-319-70347-3_3 fred moseley describes the distinction here for example.Also many marxists from what i have seen are okay with the idea of emergence that total capital is an emergent autonomous property irreducible to its constituents(personally i prefer the idea of supervenience or weak emergenece rather than strong emergence ) however the diffrence seems to be is that you think that ”capital in general” isnt merely an entity or an object but it also has cognition, intentionality,mental states ie desires,beliefs ect ami getting this right?


    1. Hi Chomes, thanks for your great question, I appreciate it.

      I’m not drawing on Moseley’s work, although I am aware of it (having debated him extensively in the past) and think his work has some admirable qualities, but also areas where I disagree. Of course, this is to be expected. I believe that I’m drawing directly from Marx’s original texts, and see my own efforts as squarely in the tradition of classical Marxism. But you are right — I depart from Marx by emphasising the “control system” nature of individual capitals. In this article I try to further develop the proposition that the labour theory of value is really (or should be) a control system theory of value. Control systems do have primitive analogues of intentional states: belief-like, desire-like and action-like (imperative). So, yes, capital is an alien cognition. Hope these comments are clarifying. Best wishes, Ian.

      P.S. The “loop closing semantics” paper I reference gets deeper into the issue of the materialist foundations of intentional states.


  3. Hi Ian. Are you saying that β€œabstract labour” and β€œvalue” will exist in a post-capacity society but that they will no longer be misused by Capital? Adam


    1. Hi Adam,
      Thanks for your question. “Abstract labour” is capital’s concept of our causal powers. Like any successful concept it does refer to an objective property of the world, yet nonetheless it doesn’t fully capture its content (e.g., the scientific concept of heat does refer to something real and independent of us, yet nonetheless our concept of heat has evolved and subtly changed along with our theory and practice). So in a hypothetical post-capitalist society (presumably where capital no longer exists, and no longer controls us) then capital’s concept “abstract labour” will have ceased to exist. However, society, while it remains within the realm of necessity, will still need some mechanism to allocate labour-power, and therefore represent it and count it. So the political economy of a post-capitalist society will also need to reify the common causal powers of social labour; probably we’ll still need some kind of “average” and fungible concept of labour-power. Whether that would be a purely quantitative representation, or something more qualitative and complex, I do not know. My guess is that any new concept will bear a family resemblance to capital’s concept of “abstract labour” (I do not subscribe to Feyerabend’s incommensurability of paradigms). I hope this somewhat philosophical answer responds to your question.
      Best wishes!


  4. Hi Ian,

    Thank you for your detailed comments. I see that we approach the problem of abstract labour from a little bit different perspective, but that there is also much common ground.

    I have tried to figure out what the heck this strange substance of abstract labour really is, on and off, for about ten years now, and although I feel I have progressed somewhat, I can’t boast that I have nailed it perfectly. Some issues still remain quite mysterious.

    I agree that Rubin’s discussion of the law of value is informative, and as his writings on the topic of history of economic thought show, his knowledge on the subject of political economy are both deep and extensive.

    What was problematic in Rubin, if my memory serves me right, was his failure to follow Marx on two crucial issues: the abstraction of labour occurs in production, not in exchange (as Rubin repeats so many times), and production is social process, not merely natural or technical matter (how anyone reading Capital can miss this point is beyond me, but if I remember correctly, Rubin makes this mistake, and surely many of his followers do).

    But I really have to reread Rubin. It’s been many years since I last read Rubin, so I don’t comment on his book for now (I plan to revisit his book more systematically later this year).

    As we both know Marx thought the distinction between abstract and concrete labour was really important. If it is a matter of strange social metaphysics without much practical consequence, all the fuss seems pretty pointless. So, I think there must a sound theoretical motivation behind the distinction for Marx.

    I think Marx’s theoretical point about the distinction between abstract and concrete labour was, among other things, to prove the theoretical ground for his conception of duality of labour process and value addition process, and as a consequence his distinction between variable capital (flows from abstract labour) and constant capital (flows from concrete labour).

    Marx defined value as labour time that is congealed to commodities. This definition of value stands as long as labour time remains the pre-eminent constraint for continuous additional production. Variable capital is “variable” because only it can add value. Constant capital, on the other hand, merely transfers the value already in place in the means of production.

    I see that this connection isn’t emphasized much in current academic discussions about abstract labour, although already Rosdolsky, quite forcefully, argues for it in Making of Marx’s Capital. A point hard to miss, if you read Grudrisse carefully in my opinion.

    My point about competition being the mechanism of equalisation of labours (while exchange is merely a consequence of the workings of this mechanism) is not new either. It was argued already by Paul Sweezy 1942 in his Theory of Capitalist Development, and I think Henryk Grossmann said pretty much the same even earlier.

    By allusion to “real competition” I refered to work of Anwar Shaikh and his co-workers (James Clifton, Willy Semmler, Dumenil etc., standard references can be found in Shaikh’s book Capitalism (2016)) that have developed empirically grounded classical marxian notion of competition without dependence on neoclassical notions of perfect competition, general equilibrium, hyperrationality, and all the rest of it.

    I think, for example, that if you follow Marx’s three-sided conception of competition (competition within capital between labour and capital, competition of capitals within industry, competition of capitals between industries), the duality of labour is manifested differently at these different levels of abstraction.

    Also, my point about capital was mainly about the need the distingius more carefully in which sense you are speaking about capital. Thus, I would agree that capital in its most general sense as capital-as-totality may be viewed as alien cognition that is able to possess certain (otherwise even pretty decent persons) people, most directly capitalists themselves.

    You wrote:
    “[T]he causal powers of human labour are β€œuniversal” in the sense of computability theory: that in principle we have the power to understand any aspect of the material world, and control it.”

    Well, this sounds like Evald Ilyenkov’s philosophy to me. I guess, as someone who takes Hegel quite seriously, you would appreciate his Dialectical Logic and concept of the ideal.

    Best wishes,


    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s