The Middle Ages

How to Read a Customs Roll: What the Grande Coutume Can and Cannot Prove

The 102,724 tuns of 1308–09 come from a tax record, not a harvest report. What the Bordeaux customs accounts actually counted, and what they cannot show.

The most quoted number in the history of Bordeaux wine is 102,724 tuns, exported in the campaign year 1308–09. It appears on museum labels, in the first paragraph of trade histories, and on this site. It comes from a tax account. Read strictly, it is the quantity of wine on which duty was reckoned at the Bordeaux customs of an English duchy, in a unit whose volume in litres is still disputed, from a series with gaps in it. It is not a harvest figure, a shipping register or a count of vineyards, and each of those differences limits what the number can be used to show.

The document

The figure comes from the accounts of the constables of Bordeaux, the duchy's chief financial officers, and from the customs books kept under them at the port. Their business was the Grande Coutume, the export duty the Plantagenet administration levied on wine leaving Gascony.

The constables were royal officers, so their accounts went to the English exchequer for audit. That is why they survive: they are in the National Archives at Kew, in series E 101 and E 364, while nearly all the records the duchy kept in Bordeaux have been lost. Robert Blackmore re-examined these accounts for his 2018 Southampton thesis, analysing more than 3,500 individual shipment entries between 1348 and 1449, and concluded that the constables' books are the best evidence available for the commercial life of Gascony. Their survival is an accident of English audit practice. Medieval Bordeaux can be counted because a clerk in Westminster had to check that a royal officer had not stolen the receipts.

Margery K. James turned the series into economic history. Her 1951 article in the Economic History Review and the posthumous Studies in the Medieval Wine Trade (1971) produced the export totals still quoted today, including those behind the chart on our English centuries page and the CSV on our data page.

What was taxed

The Grande Coutume was charged per tun. In the early 1350s it stood at 13s 4d bordelaise, then about 4s sterling, on every tun exported. Blackmore reconstructs several smaller levies from the same accounts. The issac was a transaction tax of 6s 8d on wine exchanged inside the boundaries of Bordeaux. Killage charged 10s bordelaise on a ship using the harbour for the first time. The Royan took 2½d per tun for passage out of the Gironde estuary. A customs book is the written record of these charges.

The entries are organised accordingly. Beside almost every shipment the clerks wrote one of six abbreviated tax codes: wine produced under franchise; wine that had paid both the custom and the transaction tax; wine bought from privileged estates at a reduced rate; wine from outside the city's boundaries, exempt from the transaction tax; wine from a privileged town of the Haut-Pays; and so on. The roll sorts wine by fiscal status and provenance. It says nothing about vineyard, the producer's intentions or quality. In Blackmore's price dataset, 95.6% of transactions are simply "claret", and no source records a quality grade.

Four things the roll cannot prove

Production. Port throughput includes wine that was only passing through. After 1353 this becomes impossible to ignore. An English statute that year made Bordeaux the compulsory staple for Gascon wine exports, and the privilège des vins obliged growers upriver to bring their wine down to the city. The constable's account for 1357–58 shows about 2,130 tuns grown up the Dordogne being shipped to Bordeaux before going overseas, against 83½ tuns exported directly from Libourne. The law had re-routed 96.1% of the area's exports. A rise in the Bordeaux total can reflect a change in regulation, with nothing changing in the vineyards.

The port of Bordeaux on its own. Here the two main readings of the evidence disagree. Sandrine Lavaud, the leading French specialist, presents the 102,724 tuns as the peak of traffic through the port of Bordeaux. Blackmore, working from the same accounts, treats James's totals as exports from Gascony as a whole and shows how much left from other quays. In 1306–07, 1,155 of the 93,452 tuns came from Bayonne. In the five comparable years before 1337, ports downstream of Bordeaux, Libourne excluded, shipped between 6.8% and 21.2% of the total. In the record year 1308–09, Bordeaux's rivals together, including Libourne with 9,126 tuns, accounted for 20.2%. Argument will not reconcile the two positions. Attributing each surviving entry to its port of loading would, and Blackmore's appendices have started that work. Until it is done, the defensible wording is Gascon exports recorded at the Bordeaux customs, not "Bordeaux produced".

A fixed volume. The tun was notionally 252 statutory English gallons. The usual conversion of about 900 litres goes back to Yves Renouard's 1953 study in the Annales du Midi. In a follow-up article in 1956 Renouard left the question open: he could not decide whether the fourteenth-century Bordeaux tun already held 900 litres or still only 750 to 800. Lavaud has argued that the tun was less a physical barrel than a widely used unit of account. In 1445 English merchants complained to the crown that Gascon tuns were being "made of smaller measure than the English gauge". The complaint shows that a standard was expected, and that it was not reliably kept. The familiar equivalence, 102,724 tuns ≈ 850,000 hectolitres, therefore carries an uncertainty of roughly one sixth of its own value, and should be printed with that range shown.

Ownership. Gascon names appear beside shipments because provenance set the rate of duty. They do not show that the person named owned the wine at sea. Franchise shipments in particular had to name the producer to prove the exemption. An older privilege distorts the record further: in 1214 King John exempted the burgesses of Bordeaux from the customs. Fiscal status decides who is written down and in what form. The roll is good evidence for how privilege was structured and poor evidence for who owned what.

Gaps in the series

Usable customs books survive for only about a third of the last century of Plantagenet rule. Our own series has 42 campaign years for 1305–1453, one of them an estimate. A blank year means a lost register, not a year without trade, and any chart of this material, ours included, should say so in the caption.

The accounting year was not the calendar year. It began at Michaelmas, 29 September, so a total headed "1372–73" describes the shipment of one vintage over the following six or seven months. In a sample of 2,933 recorded consignments, 54.8% left in November and December.

What the accounts do show

The same administration levied the same duty under the same rules for a century and a half. That consistency makes the accounts a strong record of relative size and direction of change: the scale of the trade in the early fourteenth century, its collapse after 1337, the partial recovery of the 1360s, the low plateau of the fifteenth century. They are also the main evidence for the fiscal arrangements the economics of Bordeaux rested on: who paid what, who was exempt, which towns held privileges.

Popular history relies on them most for total volume in litres, for where "Bordeaux" ends, and for production. On those three points they are at their weakest.