Debt Before Coinage

14 min read

Debt Before Coinage

For most of human history, people did not barter --- they ran tabs of mutual obligation. Coins appeared thousands of years later as a way to make credit portable between strangers.

What is it?

The standard economics textbook tells a story about the origin of money that goes roughly like this: in the beginning, people bartered --- a farmer with wheat met a herder with sheep, and they swapped. Barter was clumsy because it required a “double coincidence of wants” (you need what I have, and I need what you have, at the same time). Money was invented to solve this problem. First came commodity money (shells, salt, metal), then coins, then paper, then digital accounts.1

This story is almost entirely wrong.

The anthropologist David Graeber, drawing on decades of ethnographic and historical evidence, demonstrated in Debt: The First 5,000 Years that no society has ever been found whose primary mode of exchange was barter among strangers.1 What existed instead, across virtually every documented pre-monetary society, was credit --- informal systems of mutual obligation between people who knew each other. You gave me an axe today. I did not pay you. We both understood that I owed you something of roughly comparable worth, and at some unspecified future time, we would sort it out. The “debt” lived as a shared understanding, perhaps marked on a tally stick or tracked by a third party, but not settled in the moment.

Coinage --- stamped metal of guaranteed weight and purity --- appeared around 600 BCE in Lydia (modern-day Turkey), roughly seven thousand years after the earliest evidence of credit-based exchange.2 Coins were not invented to replace barter. They were invented to make credit portable between people who did not know or trust each other: soldiers, merchants operating across borders, and states collecting taxes from subjects they would never meet personally.

This reversal matters because it changes what money is. If money evolved from barter, then money is fundamentally a commodity --- a more convenient sheep. If money evolved from credit, then money is fundamentally a record of debt --- a portable IOU backed by social trust. The evidence overwhelmingly supports the second story.1

In plain terms

Imagine a village where everyone knows everyone. Nobody needs coins because everyone keeps a mental tab: “I owe the baker bread, the baker owes the carpenter shelves, the carpenter owes me a fence repair.” The tabs are the economy. Coins only become necessary when a stranger rides into town and nobody knows whether to trust them.

At a glance

How does it work?

The barter myth

The story begins with Adam Smith. In The Wealth of Nations (1776), Smith proposed a hypothetical state of affairs in which people exchanged goods directly --- “a certain propensity in human nature… to truck, barter, and exchange one thing for another.”3 He acknowledged this was a thought experiment, not a historical claim. But over the following two centuries, the thought experiment calcified into textbook fact. Generations of economics students learned that barter was the original mode of exchange and that money evolved to solve its inefficiencies.

Graeber surveyed the anthropological literature and found a striking gap: there is no ethnographic record of any society that used barter as its primary domestic exchange mechanism.1 Barter does occur, but almost exclusively between strangers or enemies --- people who have no ongoing relationship and no reason to extend trust. Within communities, a different system prevails.

Think of it like...

The barter myth is like the claim that humans naturally walk single-file in straight lines. It sounds plausible in a textbook. But watch actual humans and they walk in clusters, stop to chat, and loop back. The textbook description fits a model, not the reality.

How credit actually worked

In most pre-monetary societies, exchange operated through what anthropologists call systems of mutual obligation.1 The mechanism varied --- gift economies in the Pacific Islands, cattle-debt networks in East Africa, grain-credit systems in Mesopotamia --- but the underlying logic was consistent:

  1. Goods or services flow without immediate settlement. You give me fish today. I do not give you anything back today.
  2. An obligation is created. Both of us understand that I now owe you. The obligation may be precise (“ten measures of grain by harvest”) or approximate (“something of comparable worth, eventually”).
  3. The obligation is tracked. In small communities, social memory suffices. In larger ones, physical records appear: tally sticks, tokens, knotted cords, clay tablets.
  4. Settlement happens later, if at all. In many societies, debts were never fully settled --- they rolled over, creating an ongoing web of mutual dependence that was itself the social fabric.1

Why barter fails as a daily system

The textbook explains that barter requires a “double coincidence of wants”: you must have what I need, and I must have what you need, at the same time. This is presented as the reason money was invented.

But Graeber pointed out a deeper problem: in a community of people who know each other, the double coincidence of wants is irrelevant.1 If you need bread and I have bread, I simply give it to you --- because I know we will interact again tomorrow, and next week, and for years to come. The ongoing relationship is the credit. Barter logic only applies when there is no relationship and no expectation of future interaction. This is why barter appears historically between strangers, enemies, and trading partners from distant societies --- not between neighbours.4

Think of it like...

You do not barter with your housemates. If your housemate cooks dinner, you do not demand they hand over a precisely equivalent service before you eat. You eat, and you both vaguely understand that you will cook next week, or do the dishes, or buy the groceries. The household runs on credit, not barter. Now scale that up to a village of three hundred people, and you have a pre-monetary economy.

The arrival of coinage

The first coins --- small lumps of electrum (a gold-silver alloy) stamped with a symbol of guaranteed weight and purity --- appeared in Lydia around 600 BCE.2 Within a century, coinage spread across the Greek world and the Persian Empire.

Why then? Graeber argued that coinage emerged in the context of three concurrent developments: standing armies, long-distance trade, and taxation.1 All three involve transactions between strangers.

  • Armies need to be paid, and soldiers cannot run a tab with their commander. Coins are portable, standardised, and anonymous --- ideal for paying thousands of strangers.
  • Long-distance merchants trade with people they may never see again. A credit system requires ongoing relationships; coins do not.
  • States need to collect taxes from subjects across a wide territory. Requiring payment in stamped coin gives the state control over the unit of account and a way to verify that taxes are paid.

Coinage, in this view, did not replace credit. It supplemented credit for a specific use case: transactions between parties with no ongoing relationship. Credit continued to be the dominant mode of exchange within communities for centuries after coins appeared --- and in many ways, it still is. Your bank account, your credit card, your tab at the local shop are all credit instruments, not coins.5

Concept to explore

See credit-and-trust for how the credere chain --- the chain of belief that makes credit work --- underpins the entire modern financial system, from your bank account to interbank lending.

The cycle continues

Graeber identified a recurring historical pattern: periods dominated by credit (virtual money) alternate with periods dominated by coinage (physical money), and the transitions are often violent.1

  • 3500—800 BCE: Credit economies dominate. Mesopotamian temples run on grain-credit. Money exists as a unit of account, not as a physical object.
  • 800 BCE—600 CE: The “Axial Age” of coinage. Standing armies, empires, and slavery drive the adoption of physical currency.
  • 600—1500 CE: Return to credit. The Middle Ages in both Europe and the Islamic world are dominated by credit instruments, letters of exchange, and trust-based networks.
  • 1500—1971 CE: Return to metal-backed currency with European colonialism, gold standards, and global bullion flows.
  • 1971—present: Return to credit. After the end of the gold standard, money is once again purely virtual --- entries in ledgers backed by institutional trust.

This cyclical pattern suggests that credit is the default and coinage is the exception --- a technology adopted when social trust breaks down or when states need to project power across distances.1

Why do we use it?

Key reasons

1. Corrects a foundational misunderstanding. The barter myth shapes how people think about money, trade, and financial innovation. Replacing it with the historical record changes the questions you ask and the assumptions you carry. 2. Reveals what money actually is. If money evolved from credit rather than barter, then money is a record of social obligation, not a commodity. This explains why money “works” even when it has no intrinsic value --- because it never needed intrinsic value in the first place. 3. Clarifies modern finance. Bank accounts, credit cards, government bonds, and interbank lending are all credit instruments. Understanding that credit is the original form of money, not a modern innovation layered on top of “real” money, makes the modern financial system less mysterious.

When do we use it?

  • When someone claims that money must be “backed by something” to have value --- the credit view shows that most money in history was backed by nothing but mutual trust
  • When evaluating claims about cryptocurrency “returning to sound money” or “eliminating the need for trust” --- the historical record shows that eliminating trust is neither possible nor desirable
  • When studying the origins of inequality, since systems of debt and credit are also systems of power and obligation
  • When trying to understand financial crises, which are almost always crises of credit and trust rather than shortages of physical money

Rule of thumb

If the explanation for a financial phenomenon starts with “people naturally barter,” treat what follows with scepticism. The historical evidence points in a different direction.

How can I think about it?

The pub tab

A regular at a local pub does not pay for each drink individually. The bartender keeps a tab --- a running record of what the regular owes. At the end of the week (or the month), the tab is settled. This is credit, not barter, and it works because the bartender and the regular have an ongoing relationship built on trust.

  • The tab is the ledger of mutual obligation
  • Trust is what makes the tab possible --- the bartender believes the regular will pay
  • Cash is what a stranger uses, because the bartender does not know them well enough to extend a tab
  • Coins in Graeber’s account play the same role as the stranger’s cash: a technology for transactions without an ongoing relationship
  • If the pub goes cashless and only runs tabs, it has not invented something new --- it has returned to the oldest form of exchange

The neighbourhood babysitting circle

A group of families agrees to babysit for each other. Nobody pays cash. Instead, they track hours: if you watch my children for three hours, I owe you three hours. A simple ledger (a shared spreadsheet, a notebook on the fridge) records the credits and debits.

  • The hours are the unit of account --- not money, but a measure of obligation
  • The ledger tracks who owes whom, exactly like a Mesopotamian temple tablet
  • Settling the debt does not require money --- it requires reciprocal action within the same trust network
  • The system breaks down only if someone free-rides (takes babysitting but never provides it) or if trust collapses (families stop believing the ledger is accurate)
  • Introducing cash into this system would not improve it --- it would replace a functioning credit network with an unnecessary intermediary

Concepts to explore next

ConceptWhat it coversStatus
ledger-primacyThe ledger as the substrate from which civilization growscomplete
credit-and-trustThe credere chain underlying all financial systemsstub
money-as-social-technologyMoney as transferable credit, not commoditystub

Some cards don't exist yet

A broken link is a placeholder for future learning, not an error.

Check your understanding

Where this concept fits

Where this concept fits 

graph TD
    E[Economics] --> DBC[Debt Before Coinage]
    E --> LP[Ledger Primacy]
    E --> CT[Credit and Trust]
    E --> MST[Money as Social Technology]
    DBC --> CT
    LP --> DBC
    style DBC fill:#4a9ede,color:#fff

Related concepts:

  • ledger-primacy --- the ledger that tracks credit obligations is the substrate from which economic coordination grows
  • credit-and-trust --- the belief chain (credere) that makes credit systems function, from village tabs to interbank lending
  • money-as-social-technology --- the view that money is transferable credit, not a commodity that happens to circulate

Sources

Further reading

Resources

Footnotes

  1. Graeber, D. (2011). Debt: The First 5,000 Years. Brooklyn, NY: Melville House. See especially Chapter 2 (“The Myth of Barter”) and Chapter 3 (“Primordial Debts”). ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7 ↩8 ↩9 ↩10

  2. Seaford, R. (2004). Money and the Early Greek Mind. Cambridge: Cambridge University Press. On the emergence of coinage in Lydia and its social context. ↩ ↩2

  3. Smith, A. (1776). An Inquiry into the Nature and Causes of the Wealth of Nations. London: W. Strahan and T. Cadell. Book I, Chapters 1—4 on the division of labour and the origin of money. ↩

  4. Humphrey, C. (1985). “Barter and Economic Disintegration.” Man, 20(1), 48—72. Ethnographic evidence that barter occurs between strangers or in conditions of social breakdown, not as a baseline mode of exchange. ↩

  5. Martin, F. (2013). Money: The Unauthorised Biography. London: The Bodley Head. Chapters 1—3 on money as a social technology of transferable credit. ↩

Graph View

Backlinks