Debt by David Graeber - a review

I have heard it said that good sex can keep someone in a bad relationship; after reading and researching David Graeber’s Debt I think much the same can be said of good writing and bad ideas. This book has haunted me for the last few years; I need to exorcise this demon. When I decided to start writing essays, it was my first target, and yet after accumulating twenty thousand words of notes and months of research, I never even started the essay. This is my attempt to put a ghost to rest.

I think that there are many problems with the book. However, rather than solely dunking on Graeber for a few thousand words, I will focus on tensions that pull back and forth when I consider his writing: writing as conversational vs structured; atomised and narrow analysis versus an acknowledgement of the world as messy and chaotic; and the relative importance of posing good questions, versus providing accurate answers. Graeber’s willingness to do something very different to much of what I have read before is why it left such an impression on me (and I think so many others). I want to pull on that thread, and discuss what elements I found inspiring.

Before I continue, let me summarise the first chapter of Debt, which motivates the rest of Graeber’s enterprise. He recounts a conversation with a liberal lawyer at a party he attended, in which he tells her about his involvement with campaigns at the turn of the century against international organisations such as the WTO and the IMF. These institutions were the enforcers of debts owed by African countries to American banks and hedge funds. They have many problems: often the debts were incurred by dictators and corrupt politicians who stole the money; irresponsible lending caused a crisis when macroeconomic conditions caused interest rates to skyrocket. The IMF offered debt refinancing conditional on poor countries undergoing austerity, leading to societal breakdown, starvation and disease. Such conditions included ‘better democratic governance’, even as the IMF forced countries to adopt policies their populations disagreed with. Many of these conditions, cooked up by free-market fundamentalists in Washington, did not even work, despite the misery they inflicted. On top of all this, because the IMF would act to guarantee American hedge funds’ debts (which the hedge funds knew), it actually incentivised them to make irresponsible loans in the first place, and to continue to do so, knowing that the institutions of the West would force payment.

In response to (a hopefully shorter version of this), the attorney simply said, ‘they’d borrowed the money - surely one has to pay one’s debts?’. For me, this would trigger annoyance at the pretentious use of ‘one’. According to Graeber, it triggered him to write a book: Debt.

For Graeber, this attitude betrays a widespread ‘moral confusion’ about debt. People who don’t pay their debts are bad; at the same time, people who lend money for a living are viewed as evil. The concept of debt is used to refer to the moral obligations we owe one another, while also referring to a quantifiable sum which impersonally grows over time, may be bought, sold or transferred, and ignores interpersonal relationships. Debt is strongly tied to violence. This is true not only in its enforcement but also in its imposition. The most striking example of the latter is when France unilaterally imposed a debt on Haiti after their successful rebellion. France used warships and diplomacy to enforce this debt, threatening Haiti (a country entirely dependent on trade for supplies such as food) with an international embargo if they did not agree to the debt. This debt was calculated to be as extractive as possible, initially consisting of 5 annual payments of 30 millions francs, each of which was six times larger than Haiti’s national income. To pay even the first instalment requested taking out further loans from private French banks, incurring a ‘double debt’ beyond the primary demand. After a while, American banks, especially Citigroup’s predecessor, provided further loans, and ultimately spurred the US to invade Haiti and occupy it for 19 years, enslaving locals to build infrastructure for American business.

How can the international community view debt - the consequence of such events - as natural, inevitable, and unavoidable? Haiti is perhaps the most extreme example of debt being used to facilitate violence, but Graeber asks whether this points to a more general phenomenon of debt allowing us to ignore the moral and social relationships that ought to form the core of human life. Exploring this and related questions is the purpose of Debt. Graeber explores many stories just as interesting as Haiti’s and while I would love to expand on and engage with many of them in this essay, to do so would explode its length. As such, I will attempt to limit my scope to higher-level questions about writing, arguments and history.

The first such question is whether non-fiction ought to be clear and structured, or conversational and surprising. Years of business writing, competitive debating and mathematics proofs have led me to a love of signposting. Signposting, in writing and rhetoric, refers to clearly stating up-front what the structure of your argument is: “first I will discuss X, then show how this leads to Y, and finally address objection Z’, for example. Graeber does not do this at all. This would perhaps be tolerable in a simple linear argument, but Debt is anything but. Graeber’s writing not only has a conversational tone, but also a conversational structure. Discussions with my friends often swing from topic to topic based on vibes rather than logic, as you would expect. Graeber does the same on the page. He smoothly slides from one idea to the next, even when they are logically quite disconnected. For example, Chapter 8 is ostensibly focused on outlining Graeber’s argument that history has broadly alternated between credit money and physical money over time. Yet it starts with an unrelated anecdote about slavery. This confuses the reader, and its only purpose seems to be to provide a rhetorical or ‘vibes’ connection to the previous chapter, which discusses slavery extensively. Graeber’s insistence on having one idea slide into the next verges on obsession. It must have taken a lot of careful restructuring to achieve such transitions, and yet they detract from Debt by obfuscating the actual logical structure of his arguments.

This points to a broader tension. Is there a tradeoff between writing non-fiction in a way which is surprising, engaging and not intimidating, and writing in a way which is precise, logical and makes your argument clear? I would particularly note two different aspects of writing being ‘accessible’. In the context of writing, ‘accessible’ typically refers to it being not intimidating, lacking jargon, and generally being easy to read (in the sense that your eyes can flow over the page). Debt, generally speaking, is accessible in this sense of the word. However, another aspect of writing being accessible would be making the underlying argument accessible. To achieve this, premises and assumptions need to be laid out, definitions need to be precise, and the steps of the argument made clear. Debt is completely inaccessible from this perspective. For a specific example, Graber at one point introduces a taxonomy of different modalities of human relationships: communistic relationships (giving freely to others as they need, assuming only that others will apply the same principle to you without requiring tit-for-tat reciprocity), exchange relationships (reciprocal and typically competitive, such as buying and selling in a marketplace), and hierarchical relationships (which operate according to precedent and custom, and typically are justified on the basis of identity). There is something to this taxonomy. I recall at the time of first reading, that the idea of separating communistic giving from any expectation of direct reciprocity to be especially compelling. 

However, the framework is operationalised in such a loose way that it makes it difficult to grapple with his argument. It is unclear whether Graeber intends to describe a) the causes for people’s behaviour in relationships, even if they are unaware of them, b) how people would honestly describe their relationships, or c) how they would justify their behaviour. Furthermore, what of the role of violence and power in relationships, which is completely absent from his tripartite framework? A serf-lord relationship in feudalism is the model example for Graeber’s hierarchical relationship, and yet by Graeber’s telling, it is custom which leads the serf to pay taxes to his lord, rather than the latent threat of violence. This is particularly puzzling, given Graeber’s annoyance in previous essays at social theorists who ignore the role of violence. See the following from his previous work Utopia of Rules: “The whole arrangement is the fruit of violence and can only be maintained by the continual threat of violence […] In a case like this, what we talk about in terms of “belief” are simply the psychological techniques people develop to accommodate themselves to this reality”. See also: “This is what makes it possible, for example, for graduate students to be able to spend days in the stacks of university libraries poring over Foucault-inspired theoretical tracts about the declining importance of coercion as a factor in modern life without ever reflecting on that fact that, had they insisted on their right to enter the stacks without showing a properly stamped and validated ID, armed men would have been summoned to physically remove them, using whatever force might be required.“

Another example of the muddiness of Graber’s taxonomy is the following: what if I do altruistic actions for my friends and family, simply because that’s what I do and have always done for friends and family, based on their identity as my friends and family? Under this reading, we might fold the entire ‘communistic’ modality under ‘hierarchy’, and it would be consistent with how Graeber defines the latter. The obvious absurdity of this points to how Graeber’s definitions are lacking in precision. As a reader, the use of the word ‘hierarchy’ to describe a relationship suggests a relation of inequality in some way, and yet Graber never explicitly includes this as part of his definition. By and large, he doesn’t even offer a definition. He seems to prefer to let implicit and latent definitions for his concepts build up in his readers’ minds. Such ‘unconscious definitions’, I think, let Graber get away with logical murder. 

You can guess how I feel about this. Yet there is still something so alluring about the possibility of taking the best parts of the above, and avoiding the downsides. Consider again signposting - tell them what you’re going to say, say it, then tell them what you said - which is the hallmark of good scientific and business writing. Yet saying upfront what you’re going to say eliminates the possibility of surprise. Perhaps signposting, artfully done, might still leave space for surprise in the details of an argument. But my enjoyment and frustration with Debt forces a question: is it possible to write a well-argued and accessible essay without signposting, while still leaving the reader in a position to understand how your argument hangs together? More generally, is it possible to retain an attractive conversational tone, while producing a logically accessible work?

I will now turn to a different challenge that Debt poses. Reviews have correctly identified Debt as being a disciplinary move disputing economics’ ascendency over anthropology, and a politically-motivated history of anarchism against both big-state liberalism and monopolistic capitalism. But it is also a book which aims to be a ‘big book’ answering ‘big questions’, and as such, challenges the narrowness of academic research. As Graeber explicitly describes: “a big book, asking big questions, meant to be read widely and spark public debate, but at the same time, without any sacrifice of scholarly rigor”. Beyond this quote, Graeber’s debt clearly aims to formulate and apply a theory which applies across all of human history. Is it possible to do this while still being true? 

Academic work typically focuses on narrow and atomised phenomena. This narrowness can occur in several ways: focusing on a particular dataset, on a particularly defined phenomenon of interest (whether quantitative or qualitative), in a particular geographical, cultural, or temporal context. Conclusions from such studies are also often limited to this context - the ‘external validity’ of a study refers to what extent its conclusions can be applied beyond the context in which it was performed. There are good reasons for narrowing scope in this way: appropriate application of statistical methodologies can provide confidence that spurious conclusions have not been reached, and isolating phenomena allows them to be studied. At the same time, such focus can eventually become blinkers. Graeber argues that the discipline of economics has suffered in exactly this way, reducing complex motivations for people’s behaviour to pure self-interest. That argument is not especially original; Graeber goes further in arguing that this is not only an academic mistake, but has actually infected how we think in our society. From this perspective, the breadth of Debt’s scope is not only a challenge to academic norms, but also to ingrained assumptions about how we understand the world. The problem is that Graeber can’t seem to decide whether he wants the world to be so complex that it resists understanding from economics’ simplistic tools, or simple enough to admit a simplistic narrative. In one section, he lyrically waxes, ‘This particular maze of mirrors is so complex and dazzling that it’s extraordinarily difficult to discern the starting point - that is, what, precisely, is being reflected back and forth.’ In another, he argues - absurdly - that the structure of history is sufficiently simple, that history across the entire globe can be broken down into alternating periods of commodity and credit money, thousands or hundreds of years long. 

The most irksome contradiction in this vein is that Debt actually mirrors the destructive pessimism of economics. Graeber repeatedly criticises economics as being a method by which bad things become inevitable and therefore justified: “This is what the use of equations so often does: make it seem perfectly natural to assume that, if the price of silver in China is twice what it is in Seville, and inhabitants of Seville are capable of getting their hands on large quantities of silver and transporting it to China, then clearly they will, even if doing so requires the destruction of entire civilizations. Or if there is a demand for sugar in England, and enslaving millions is the easiest way to acquire labor to produce it, then it is inevitable that some will enslave them.” However, Debt itself indicates that extractive and coercive debts, far from being a modern phenomenon, have always existed - Graber gives examples from Mesopotamia, ancient Egypt and China, all the way through to human times. What, exactly, are we meant to conclude from this? His narrative seems to suggest that harmful debt is part-and-parcel of human civilisation. From this perspective, economics and Graeber’s anthropology are two sides of the same coin - whereas economics gleefully justifies coercive debt while insisting that it is inevitable, Graeber criticises it, while also suggesting its inevitability. Graeber presents a few examples of small tribal societies which arrange themselves differently and (according to his reading) avoid such issues, but they are hardly encouraging, compared to the long historical track record. He calls for periodic debt forgiveness, and certainly we could point to Debt as helping trigger the discourse that has now led to actual debt cancellations (especially student debt), but such ad hoc events hardly constitute structural changes. 

In the same blog post where Graeber talked about having written a ‘big book’, he admitted, ‘History will judge whether it’s still possible to pull this sort of thing off (let alone whether I’m the person who will be able to do it.)’ As is clear from the above, I do not think that he was the person who was able to do it. Thomas Piketty’s Capital represents a more compelling, if less expansive, attempt to answer a truly broad question with scholarly rigour. In particular, I agree with Mark Webbs’ review of Debt in Jacobin: “Graeber stays almost entirely within the domain of “moral universes” and discourse. We don’t get a sense of just how the moral economy ofMerrie England was undermined, except that the powers-that-were didn’t get it, didn’t like it, and imposed their own morality somehow.” Graeber refuses to engage with the world as-it-was, and instead focuses on the mythology and discourse of and about the past. This prevents him from actually telling any concrete history at all. 

Having said all this, I think that writing a ‘big book’ (in the sense of attempting to explain a wide variety of human phenomena) opens you up to criticism, because you necessarily need to flatten the truth in order to conform it to your model. I think that it is valuable. I will look for better iterations.

Finally: what is the importance of accurately reporting the facts, relative to prompting important questions? One word which came up a few times when reading friendly reviews of Debt was the word generative: the book may not be (at all) accurate, but it does prompt the reader to think of the world in a different way, and to come up with new ideas. In some views, this is the highest aim of anarchist writing - to pierce the veil of assumed wisdom about how the world works - and so perhaps Debt succeeds by this measure alone. When I conclude this essay, I will describe one way in which Debt changed my life. But before that, I want to elaborate on just how deeply disappointing *Debt *is as a work of scholarship. It is inaccurate in so many ways. There has been a lot of criticism about Graeber’s conclusions about the modern economy - I recommend this article from Crooked Timber and this review in Jacobin. Those are easy pickings, but I found that even when I waded into Graeber’s discussions about philosophy, discourse, mythology, and language, evidence did not support his conclusions. Here are three examples where when I looked into the footnotes and supporting evidence for a claim, it didn’t stack up.

Firstly, when discussing philosophies which ‘adopted the terms of debate’ from the cynical realists who focused on profit and advantage, Graeber uses the example of Mohism, which was born during China’s Warring States period. He points out how Mohism uses the notion of li (which derived from profit, but came to mean ‘benefit’ in a more general sense) to argue that war was unprofitable even for the victor, due to the cost of war on states’ populations. In Graeber’s view, the use of li betrayed that Mohism ultimately relied on concepts from the impersonal market. However, Graeber’s argument is inconsistent with other aspects of Mohism. For example, Mohism argued that disorder arises because of individuals’ conflicting ethics, which motivated Mohism to try to build social structures that promoted alignment between people’s personal ethics. If Mohism had truly ‘adopted the terms of debate’ from cynical realists, we would expect its explanation of disorder to be the result of people acting in their self-interest, instead.

Secondly, Graeber later wants to chart the progression of the notion of freedom from, in pre-ancient times, being defined as ‘being able to have social relationships’ (prioritising communistic relations and relationships with others) to, in modern times, ‘having domination over one’s own person and property’ (prioritising exchange relations and self-independence). He particularly focuses on the Roman concept of dominum. He argues that in early Roman times,‘to be free meant to be able to make friends, to keep promises, to live within a community of equals’. He relies on two arguments. Firstly, he leans heavily on Patterson’s definition of slavery in Slavery & Social Death as ‘social death’, which asserts that slavery is uniquely characterised by the removal of all social relationships. If this were true, then since freedom was ‘not slavery’, freedom is the ability to form social relationships. On this, I agree with a reviewer of Slavery & Social Death, that while this seems like a useful way to understand slavery, it is too absolutist, and there are plenty of examples of slaves gaining various forms of social status, and of having social relationships with one another - especially in Rome. His second argument is that ‘free’ and ‘friend’ share etymological roots because to be free means to be able to make friends. I looked into his actual source for this (which was very obscure), and it seems, according to that source, that ‘free’ is derived from expressions of familiarity between people in an in-group. Rather than expressing communistic relations, this is more suggestive of hierarchical relations: to be ‘free’ meant to be on top of the social hierarchy, or at least not on its lowest rung.

For a third and final example, Graeber seemingly misrepresents Plato. Graeber argues that Socrates (in one of Plato’s dialogues), when refuting a counterpart’s argument that justice is merely pretext for the powerful to seek their self-interest, makes the following argument: “it’s only the existence of money […] that allows us to imagine that words like ‘power’ and ‘interest’ refer to universal realities that can be pursued in their own right [proposition A], let alone that all pursuits are really ultimately the pursuit of power, advantage, or self-interest [proposition B]”. I have added in proposition A and proposition B to make the distinction between these two propositions clear. Note that Graeber’s argument relies on Socrates arguing against proposition A, because if so, it is an ancient philosopher explicitly making the connection between money and the validity of self-interest as a pursuit. As far as I can tell, Socrates does not state proposition A. Socrates does argue against proposition B, but actually himself explicitly separates out ‘the art of pay’ from ‘the art of governing’ as separate pursuits in doing so - supporting the exact distinction that Graeber says he argues against.

There are several other severe criticisms that are validly levelled at Debt: that Graeber states his case with a level of confidence and arrogance which could not possibly attend any grand theory based on historical evidence over the last 5000 years; that Graeber cherrypicks his way through history; and that he commits the ‘genetic fallacy’, believing that understanding the long-term history of a concept is necessary and sufficient to understand its current state. 

The most disappointing problem, however, is as follows.

Consider one of the fact-checks above. Suppose that I want to check whether the fact-check actually matters - what role does this particular example, or argument, play in Graeber’s overall argument? You would not be able to tell.

I really tried very hard to map out how Graeber’s web of anecdotes and arguments depend on one another. I wanted to see what parts of his book amounted to asides (interesting, but not essential) and which parts underpinned the overall arguments of his book. I found this task impossible and exhausting - although you can read some very long notes on this here if you’d like to relive my descent into madness. 

This is completely astonishing to me. My disciplinary training is in mathematics, which requires logical accuracy, and science, which aims at clearly stating and testing falsifiable hypotheses. I do not know what standards of argument are required in areas of humanities such as anthropology, but I would expect them to be better than this.

So, this has been quite a negative review so far. Why did I continue reading Debt, researching it, and then writing this essay about it? The simple answer is that some element of Graeber’s writing did change my life. I had long understood the world in terms of reciprocity, acting towards others in the way I would want them to act towards me. However, there are several problems with this. Firstly, there is the psychological aspect - do I need to keep a ledger in my head of our respective contributions and ways of acting, for every relationship, if only to make sure that I am keeping my side of the reciprocal commitment? Secondly, and more fundamentally, this approach is defeated by relationships in which reciprocity is not possible, such as that between parent and child, or senior and junior colleague, or in cases of extreme generosity or kindness. The common sentiment in such cases is to ‘pay it [kindness, generosity, etc] forward’, but I agree with Graeber that this merely tries to reformat such cases into the ledger/reciprocity framework, when it clearly no longer fits. 

I found Graeber’s exhortation to instead apply the communistic principle - give to others as you can, and take what you need - in relationships to be beneficial in my own life. As a result, I am less anxious about failing to uphold ‘my side of the ledger’ in everyday relationships, more mindful of opportunities to be generous without expectation of return, and more secure in receiving benefits from others that I am not well-placed to reciprocate. 

Moving from the personal to the political, while my close reading of it has severely undermined my faith in any of its overall arguments, Debt at the very least has made me aware of how quantifiable descriptions of our world (including but not restricted to debts) can operate to quietly cover up important qualitative aspects of that same world. These changes in my personality, and my outlook on the world, persist despite the weakness of Graeber’s wider arguments. Compare this experience to when I recently read Grayling’s History of Western Philosophy. Although it was precise and technical in its summary of philosophy, it did not have any kind of the same effect on me.

I started this essay with a trite joke comparing Debt to a relationship. Oddly, my feelings towards it are similarly bittersweet. I was inspired by how it challenged my thinking, and wanted to more deeply engage with it by researching it further. Yet this very research betrayed the thinness of its arguments, and I ultimately spent a lot of time and effort debunking it. I am left with a sense of lost potential - there was surely a better version of Debt that could’ve been written - and inspiration that I myself could write in such a way one day.

Epilogue

Consider the following quote, which is a good articulation of the worldview that Debt encourages:

Freuchen [the author of a book about the Inuit] tells how one day, after coming home hungry from an unsuccessful walrus-hunting expedition, he found one of the successful hunters dropping off several hundred pounds of meat. He thanked him profusely. The man objected indignantly:

“Up in our country we are human!” said the hunter. “And since we are human we help each other. We don’t like to hear anybody say thanks for that. What I get today you may get tomorrow. Up here we say that by gifts one makes slaves and by whips one makes dogs.”

The last line is something of an anthropological classic, and similar statements about the refusal to calculate credits and debits can be found through the anthropological literature on egalitarian hunting societies. Rather than seeing himself as human because he could make economic calculations, the hunter insisted that being truly hu­man meant refusing to make such calculations, refusing to measure or remember who had given what to whom, for the precise reason that doing so would inevitably create a world where we began “comparing power with power, measuring, calculating” and reducing each other to slaves or dogs through debt.

It’s not that he, like untold millions of similar egalitarian spirits throughout history, was unaware that humans have a propensity to calculate. If he wasn’t aware of it, he could not have said what he did. Of course we have a propensity to calculate. We have all sorts of propensities. In any real-life situation, we have propensities that drive us in several different contradictory directions simultaneously. No one is more real than any other. The real question is which we take as the foundation of our humanity, and therefore, make the basis of our civilization.