Double-Entry Bookkeeping and the Merchants of Venice

Suppose you are a merchant in fourteenth-century Florence. You have partners who have put money into your company, agents buying wool in England and selling cloth in Naples, debts owed to you by customers in three cities and debts you owe to suppliers in two more. Some of your accounts are in florins, others in local money. At any moment you would like to know a simple thing: how is the business doing, and whose money is where?

Answering that question is much harder than it sounds. The technique Italian merchants developed to answer it, double-entry bookkeeping, turned out to be one of the most durable inventions of the late Middle Ages.

Numbers that could be calculated

One precondition was arithmetic. Roman numerals are fine for recording a number but clumsy for adding long columns of them. In 1202, Leonardo of Pisa, a merchant’s son who had learned mathematics in North Africa, published the Liber Abaci, which taught Hindu-Arabic numerals and their use in commercial calculation: converting currencies, dividing shares among partners, working out interest.

Over the following century, Italian cities developed schools, often called abacus schools, where merchants’ sons learned these methods. A generation of clerks grew up able to calculate quickly on paper. That made a new kind of record-keeping possible.

The idea of the double entry

The principle is easy to state. Every transaction affects at least two accounts, and it is entered in both: as a debit in one and as a credit in another, for the same amount. If you sell cloth to a customer on credit, the customer’s account is debited, because they now owe you, and the cloth account is credited, because you have parted with goods. If you pay a supplier in coin, the supplier’s account is debited and the cash account credited.

Because every entry has an equal and opposite partner, the totals of all debits and all credits must be equal. If they are not, there is an error somewhere. This built-in check, the trial balance, is one of the great advantages of the method. It does not prove that the books are honest, but it catches a large share of mistakes.

Give and have

Italian merchants wrote their ledgers in a form that made the two sides visible at a glance. In the Venetian style, each account occupied two facing pages. The left-hand page recorded what the account owed, under the heading dare, “to give”; the right-hand page recorded what was owed to it, under avere, “to have”. Every entry on one side of one account was matched by an entry on the opposite side of another, with a cross-reference to the page where its partner could be found.

The words debit and credit, which English took from Latin, preserve the same idea: a debit is what is owed, a credit is what is trusted or believed. Anyone checking the books could follow each entry from one account to its partner and back. The ledger became, in effect, a web of cross-checked promises.

The method also introduced accounts for things that are not physical at all. The owners’ stake in the business, the capital, has its own account. So do expenses, and so does the account that gathers the gains and losses of the business over a period. By closing those accounts at the end of a year, a merchant could see, in one figure, whether the enterprise had grown or shrunk and how much each partner’s share was now worth.

Florence, Genoa and the merchant of Prato

No single person invented double entry. Historians find its elements developing gradually in Tuscan ledgers from around 1300, in the accounts of trading companies that had branches across Europe. The ledgers of the commune of Genoa from 1340 are often cited as the earliest surviving example of the complete system applied consistently.

The richest window into this world is the archive of Francesco di Marco Datini, a merchant of Prato who died in 1410. He left instructions that his papers be preserved, and they were: hundreds of account books and well over a hundred thousand letters, now the subject of a vast literature. Iris Origo’s classic study, The Merchant of Prato (1957), uses them to reconstruct the daily life of a businessman obsessed with his accounts. Datini’s books show the method in use across a network of branches from Avignon to Barcelona.

Venice and the friar’s textbook

Venetian merchants had their own refined version of the method, sometimes called the Venetian style. It might have stayed a craft secret passed from master to apprentice, had it not been for a mathematician and Franciscan friar named Luca Pacioli.

In 1494 Pacioli published in Venice a large mathematical encyclopaedia, the Summa de arithmetica, geometria, proportioni et proportionalita. Among its many parts was a short treatise on bookkeeping, Particularis de computis et scripturis, describing the Venetian method in a few dozen brief chapters. It was not the first written account: Benedetto Cotrugli, a merchant from Ragusa, today’s Dubrovnik, had described double entry in a book on trade completed in 1458, but that work was not printed until 1573. Pacioli’s had the advantage of the printing press, which had reached Venice only a generation earlier.

Three books and a beginning

Pacioli describes a merchant’s records in three books. The memorandum, or waste book, records every transaction as it happens, in plain words, by anyone in the household. The journal rewrites each entry in a standard form, stating which account is debited and which credited. The ledger then posts each journal entry to the relevant accounts, each with its own page. He advises the merchant to begin with a full inventory of everything he owns and owes, and he explains how to close the books and carry balances forward.

Pacioli was a friend of Leonardo da Vinci, who later illustrated his book on proportion, and his fame rests on more than accounting. But his bookkeeping chapters had the widest influence. Within a century there were adaptations and translations in Dutch, English, French and German, and the method spread with European commerce across the world.

Did double entry make capitalism?

In the late eighteenth century, Goethe had a character in his novel Wilhelm Meister’s Apprenticeship praise double entry as one of the finest inventions of the human mind. In the early twentieth century, the German economist Werner Sombart made a much bigger claim: that double entry, by making it possible to see a business as an abstract sum of capital, was essential to the rise of capitalism itself.

Later historians were sceptical. In a well-known article of 1949, the accounting historian Basil Yamey pointed out that many merchants kept their books in ways that did not use the method’s full analytical power, and that large businesses were run successfully without it. The truth probably lies between the two positions. Double entry was not a magic key, but it gave merchants, and later states and corporations, a common language for describing who owned what, and a discipline for checking it. The historian Jacob Soll, in The Reckoning (2014), argues that the fate of whole states has depended on whether their rulers were willing to keep honest books and let others see them.

From two entries to a shared record

There is a limit built into double entry. It is a private system. Each merchant keeps their own books, and two merchants who trade with each other each record the transaction separately. If their records disagree, there is no single authoritative version, only two ledgers and an argument. Banks, auditors and courts developed in part to settle such arguments.

In 2005 the cryptographer Ian Grigg wrote about what he called triple-entry accounting: alongside the two parties’ own books, a digitally signed receipt that both share, so that the record of the transaction itself cannot be disputed. A few years later, the Bitcoin design proposed a different but related idea: a single ledger, copied by every participant, that nobody can alter alone. Whatever one thinks of such systems, they are clearly descendants of the Venetian ledger. They are attempts to solve, with cryptography, the oldest problem in bookkeeping: getting people who do not fully trust each other to agree on one set of books.