IN PLAIN TERMSEconomics textbooks open with barter: money was invented to spare people the trouble of swapping chickens for shoes. Graeber, an anthropologist, points out that nobody has ever found such a society. People ran tabs long before they had coins, and a great deal follows from having the order backwards.
Every argument about money rests, usually silently, on an origin story. If money appeared to make barter less awkward, it is a technology for clearing trades, and the interesting questions are about its quantity and its issuer. If it appeared as a way of recording obligations that already existed, money is a ledger, and the interesting questions are who writes the entries, who enforces them, and who is permitted to strike them out. This is the most widely read attack on the first story, and it has become the standard citation for anyone asserting that money is credit, including a great many who have not checked what it establishes. Its central historical claim is strong enough to move the burden of proof; the reach it is credited with is considerably wider than the evidence it puts on the table.
The sequence in every textbook — barter first, money invented to relieve it, credit built on top — is offered as history and has never been observed. The charge is not that it is unproven, but that anthropologists have looked for two centuries and keep finding the opposite: obligations tallied, carried, settled later, in societies with no coin.
The positive case runs on Mesopotamia, where loans, rents and fines were reckoned in silver by temple and palace scribes some two thousand years before anyone struck a coin, and where most of that silver never moved. The debts circulated; the metal sat still. Barter does appear in the record, but in the wrong place for the textbook: between strangers with no expectation of meeting again, and in the wreckage of currency systems that have collapsed. It is downstream of money rather than upstream of it.
The negative half of the case is close to unanswerable, and it is why the book landed. The positive half is weaker than its reception suggests. The deepest evidence for credit-before-coin comes from precisely those societies that kept administrative records — clay tablets survive, a handshake in a village market does not — so what the Mesopotamian material shows beyond argument is that one large bureaucratic economy ran on credit. That it shows what happened everywhere else is an inference, pressed harder than the evidence carries.
There is a second softness. Those debts were reckoned in weights of silver and measures of barley, which is to say in commodity quantities, and that concedes more to the account being demolished than the demolition admits. Economists, meanwhile, use barter as an expository device rather than as a claim about the past, so part of the attack lands where few of them are standing — though the device does silent work in the theory built on top of it, which is why dislodging it matters at all.
The centre of the book is not economics but the authority of one sentence: that a person must pay their debts, held with a force people extend to no other contract. The explanation offered is that a debt is a promise between equals turned into a number, and that the arithmetic hides both the equality and the violence needed to collect.
Underneath sits a taxonomy of three moral registers running simultaneously in any society. Everyday communism allocates by need and is the unremarked baseline of household, workplace and street. Exchange assumes equivalence, and therefore allows a relation to be squared and ended. Hierarchy runs on precedent rather than equivalence. Debt is exchange held in suspense, a relation between equals frozen until the account clears, and it is the only one of the three in which failing to clear can convert a person into property.
The load is carried by the claim that quantification requires violence: a relation can be stated as a figure only once the person has been prised out of the context that gives them their obligations, and historically that prising was done by force. The slave raid, the peon, the child pledged against a bad harvest are the cases. This is the most genuinely anthropological argument in the book, it depends on none of the history that follows, and it is what will still be standing when the rest is disputed.
Its weakness is an elision the prose performs repeatedly. The moral charge of debt talk and the economic function of credit are handled as one object, so that a peon and a pension fund arrive under the same word. Since no reader will defend debt bondage, the horror travels along the word to sovereign bonds and student loans, which are not the same relation and are nowhere shown to be. The three registers, meanwhile, are offered as description and never put to a case that could embarrass them.
The historical spine is a cycle: ages of coin and bullion coincide with empire, standing armies and mass slavery, while ages of credit coincide with institutions that restrain creditors, from Mesopotamian clean slates to medieval prohibitions on usury. The claim is that this is no coincidence: metal is what an army in hostile country needs, and credit is what a settled society can afford.
The mechanism offered is a complex of military supply, coinage and slavery, and it turns on why credit fails at distance: credit requires that the parties expect to meet again. So a state provisioning an army strikes metal, hands it to the troops, and then demands that same metal back in tax, which obliges everyone within reach of the soldiers to sell them something. Markets, on this reading, are a by-product of war finance rather than a spontaneous order that states later discovered and taxed.
This is the most ambitious stretch and the most contested, and it is what makes the book a history rather than a pamphlet. Specialists have gone after the periodisation hardest: a bracket running from roughly 800 BC to 600 AD is asked to hold China, India and the Mediterranean inside one explanation, and the medieval turn back to credit understates how much coin stayed in circulation. The strongest element is not the cycle but a smaller claim lodged inside it, that what makes a population want a particular coin is the tax demand denominated in it rather than any property of the metal. That has evidence behind it and needs none of the periodisation to stand.
The weakest is the extension to the present. Four ages are read off centuries; the fifth is read off a single administrative act in 1971, after which the reader is told that the protections proper to an age of credit have not been built. That is a prediction wearing the costume of a period, and it is where the pattern stops explaining the evidence and starts assigning it.
The last stretch turns from description to advocacy. If the obligation to repay is a settlement rather than a moral absolute, it can be refused, and what follows is a case for a jubilee: broad cancellation of international and consumer debt, alongside the observation that the lending class has been remarkably good at not absorbing the losses its own risk-taking earns.
The historical warrant is the strongest part of the proposal. Clean slates were ordinary administrative practice in Mesopotamia, the jubilee is written into Leviticus, and periodic cancellation is presented as one of the mechanisms that made an age of credit survivable rather than as a fantasy imported from outside economics. Cancellation, on this account, is not a rupture in a system that has always run on inviolable contracts. It is the restoration of a device that system has used repeatedly and lately forgotten.
What the argument never does is meet its own obvious objection. Every debt is somebody's asset, and in an economy where retirement and insurance are held as claims on debt, a general cancellation is a transfer whose winners and losers are not the classes the preceding hundreds of pages have described. The objection is acknowledged in passing and left there: no account of which debts, of who absorbs the loss, of what stops the same structure reassembling within a decade. The method that produced the diagnosis is simply not extended to the remedy.
That matters less than it sounds, because the programme is not what the book proves. The history establishes something narrower and far more durable, that the terms of debt have been rewritten, suspended and cancelled by authorities of every description, and that is enough on its own to defeat the claim that repayment is a moral fact standing outside politics. Readers arriving for the jubilee will find an epilogue; readers arriving for the argument that made a jubilee thinkable will find it finished long before.
Less than its reputation implies, on the point it is famous for. The credit alternative to the barter origin was already in print with Georg Friedrich Knapp in 1905 and stated flatly by Alfred Mitchell-Innes in 1913, then pressed again by Keynes and Polanyi; Graeber credits them and his readers mostly do not. What the book adds there is reach, putting the ethnographic record and the cuneiform material in one place at a length that made the case impossible to ignore outside economics. The genuinely new work sits elsewhere and is quoted far less: the explanation of why debt talk carries a moral charge no other contract carries, located in the act of quantification and in what quantification has historically required. That argument is his own, it survives every dispute about the periodisation, and it is the part worth taking.
Both read monetary outcomes as decisions someone made rather than weather the authorities endured — Friedman from the Federal Reserve's own deliberations, Graeber from who benefits when a debt is enforced or forgiven. What they agree on is that there is an agent to name.
SCOPE · on agency, not on interest · Inflation is chosen, not sufferedHayek's money is a market-selected commodity, chosen for saleability before any credit relation exists. Graeber's evidence is that credit reckoning and enforceable obligation predate coinage by millennia, and that the barter origin story has no support in the record.
SCOPE · on origins · Money before creditGraeber does not answer Hayek so much as refuse the terms. He reads monetary arrangements as expressions of political obligation, so a dispute about which party should issue treats as technical what he takes to be a settlement between creditors and everyone else.
SCOPE · on whether issuance is the question · Money without a monopolyThe sharpest version of the same disagreement. Ammous builds from a commodity selected for hardness; Graeber from ledgers of obligation that existed long before metal was weighed. Neither account absorbs the other's evidence without giving up its own foundation.
SCOPE · on origins · Money before creditRead the opening third and the passage laying out the three moral registers; both are better than any summary of them, this one included. What follows is an absorbing and unevenly sourced world history that mostly re-illustrates what those have already established. Five hundred pages carrying three arguments and a programme is a ratio the prose is good enough to conceal, right up to the moment one tries to state what the long middle proved. Anyone who wants only the demolition of barter can have it from Mitchell-Innes in twenty pages.
Anyone who repeats that money is credit without having read the case for it, and anyone on the other side who assumes the barter story is safe because it is in the textbooks. Also readers of monetary economics who have never had the moral question put to them at all: why this one obligation, alone among contracts, is felt as a fact about the universe rather than a term someone agreed.
The material on honour, slavery and the Mediterranean, the treatment of the great religions and their vocabulary of debt, the human economies of Africa and Madagascar, and the financial-crisis framing the updated edition sharpens. Left out for a different reason: any verdict on whether competing issuers would beat a single legal monopoly. The book explains how a money comes to be taken and never asks which arrangement serves people better, and its evident warmth toward credit monies arising outside any state is never turned into a claim.
This assessment describes what David Graeber argues in Updated and expanded edition, Melville House, Brooklyn, 2014, and is derived from that edition. It is a position taken about the argument — not a recommendation from someone who read the book for pleasure and enjoyed it.
https://summarylibrary.com/debt-the-first-5000-years · summary ref G-2011-01 · updated 2026-08-22 · the summary is not the book.