The Great Dish from the Mildenhall Treasure

The largest and most heavily decorated piece in the Mildenhall Treasure, a hoard of thirty-four Roman silver vessels struck by a ploughman's blade in a Suffolk field in January 1942. A mask of Oceanus stares from the centre of the dish, ringed by sea nymphs and, in the widest band, a full Bacchic procession complete with musicians. The dish was kept as undeclared household silver for over four years, at times doing duty as a fruit bowl, before being reported to the British Museum in 1946 and declared treasure trove.

The documented record, drawn from the coroner's inquest and from British Museum correspondence, corrects a persistent popular myth: the reward was split equally between the finder, Gordon Butcher, and his employer, Sydney Ford, at 1,000 pounds each, not unevenly as is often repeated. Roald Dahl's famous fictionalised account of the discovery, published in America in 1947 and later reworked into a well-known illustrated book, took real liberties with both men's characters; Dahl said so himself, and this record treats his version as reception history, not as a source for what actually happened.

Where the dish was made, exactly when its owner buried it, and why, remain open. The British Museum commits only to 'the general Mediterranean region' for its manufacture; the usual reading that it was hidden amid the instability of the end of Roman Britain is an argued interpretation, not a fact fixed by any evidence recovered with the hoard itself. The precise spot where Butcher's plough struck it has never been found again.

What it is

A shallow silver serving dish 605 millimetres across and weighing 8.256 kilograms, the largest and most heavily decorated object in the Mildenhall Treasure, a hoard of thirty-four pieces of late Roman silver tableware found in Suffolk in 1942. The decoration is worked in low relief and engraved line across the whole face of the dish, organised in three concentric zones.

At the centre a staring mask represents Oceanus, the personification of the ocean, with dolphins rising from his hair and a beard rendered as fronds of seaweed. Around him, inside a border of scallop shells, an inner frieze shows Nereids, sea nymphs, riding a menagerie of marine creatures: a sea horse, a triton, a sea stag and a ketos, a dragon-like sea monster. The outer and widest zone carries a band of Bacchic revelry - Bacchus, Hercules staggering drunkenly and held up by two satyrs, Pan, maenads and dancers, several playing instruments identifiable in the relief as cymbals, panpipes, the aulos, a tambourine and a frame drum.

Nothing else in the hoard approaches it in scale, weight or the density of its imagery. Every study of the object treats it as a single deliberate statement piece rather than an ordinary serving vessel, though its size and slightly domed profile are also consistent with actual use in serving food at a formal table.

The hoard it belongs to

The Great Dish is one of thirty-four silver objects recovered together and registered as a single find, the British Museum's registration numbers running 1946,1007.1 to .34. The dish itself carries the first number in that sequence. The coroner's inquest summary of the find, read into the record on 1 July 1946, described the assemblage as two large silver dishes, four large flanged bowls with beaded rims, two smaller similar bowls, one large flanged bowl with a niello rim and cover, two shallow dishes with beaded rims, one large fluted hanging bowl, two small wine cups on pedestals, five spoons with dolphin handles and eight long-handled spoons.

The second large dish, plain and undecorated, is a companion piece to the Great Dish in scale but carries none of its imagery; the two should not be confused. Every claim in this record concerns the Great Dish specifically. Facts about the pedestalled platters, the flanged bowls, the spoons or the hanging bowl are given only where they illuminate the hoard's wider context, and are labelled as such.

One feature of the hoard is itself informative: it contains no coins. A group of thirteen siliquae published in the Numismatic Chronicle in 1942 has been floated as a possible related find, but no documentation survives connecting them to Mildenhall, and the British Museum's own archive search failed to identify who submitted them. The likelihood is that they are unconnected. Without coin evidence, the hoard cannot be dated by association in the way many Roman hoards are, and its date rests instead on the style of the vessels themselves.

How it was found

In January 1942 the ploughman Gordon Butcher was working a field near West Row, Mildenhall, Suffolk, for the engineering contractor Sydney Ford. The field, Ordnance Survey plot 1673, was under the tenancy of Fred Rolfe, whose landlord was Mrs Sophie Aves. Butcher had been asked to plough deeper than usual, to a depth that struck the hoard roughly 254 millimetres (ten inches) below the surface, because the field was being prepared for a root crop. At his own testimony to the 1946 inquest, Butcher confirmed he found the treasure in January 1942, that the field belonged to Rolfe, and that he and Ford dug the objects out together before taking them to Ford's house.

The pieces were black with corrosion. Ford believed at first that they were pewter, or possibly lead. He kept them at his home, Grove Villa, West Row, and spent what he described at the inquest as nearly two years cleaning the largest tray alone, working the crust off piece by piece. He then arranged a number of objects, including the Great Dish, on the sideboard in his front room, where family photographs from around 1944 show them on open display next to the fireplace.

Sydney Ford's grandson, also named Syd Ford, told the BBC in a 2002 episode of Meet the Ancestors that the Great Dish served as a fruit bowl at Christmas, piled with apples, oranges, pears and nuts, and that one of the silver spoons from the hoard was used at the family table for breakfast and dinner. This is the specific, sourced version of the story: an object of major art-historical importance handled as ordinary household silver for years. No source consulted for this record substantiates the more colourful claim, sometimes repeated online, that pieces were used as ashtrays; that detail is dropped here for lack of a citation.

Four years of silence and the 1946 inquest

Ford did not report the find. His household had a longstanding acquaintance with Hugh Alderson Fawcett, a Buckinghamshire general practitioner and amateur antiquarian who shared Ford's interest in flint tools and had visited the family through the war years without being shown the treasure. At Easter 1946, probably 22 April, Ford finally showed Fawcett the hoard, by then split between two rooms of his house. Fawcett urged him to report it and, over the following weeks, arranged for the British Museum to examine a handful of pieces Ford was willing to part with. Christopher Hawkes and Thomas Kendrick of the Museum's Department of British and Medieval Antiquities, and Harold Plenderleith of the Museum's research laboratory, confirmed the metal was silver, not pewter as Ford still believed.

Ford remained reluctant, writing to Fawcett on 13 June 1946 that he would 'rather put them back where they came from than part with them on these terms' and that he did not feel he had 'committed a crime by picking up something of value off my land'. Once the British Museum forwarded him the official treasure trove regulations, however, his position changed abruptly. On 21 June 1946 he reported the find to the police at Mildenhall, who seized the hoard the same day; his covering letter to Fawcett read, in full: 'I reported the find to the police at Mildenhall on the 21st inst, who promptly came along and pinched the lot.'

The coroner's inquest was held at Mildenhall police station on Monday 1 July 1946 at three in the afternoon, before a jury of twelve local men chaired by the bank manager Robert Pizzey. Ford and Butcher each gave evidence, and each admitted that an earlier statement they had given police on the day of seizure had been made 'to mislead others': Ford had initially claimed to have found the treasure alone, on land he co-owned with his brother, and dated the discovery to January 1943; Butcher's first statement, given after Ford briefed him, matched this false account. Under oath at the inquest both men corrected the record: the true finder was Butcher, the true date was January 1942, and the true landowner was Rolfe, not Ford. The jury declared the hoard treasure trove, vesting it in the Crown.

The reward: what the record actually shows

Popular retellings of this story, including some that circulate widely online, describe Ford walking away with most or all of the reward while Butcher, the actual finder, received a token sum. The documentary record does not support that version. British Museum correspondence examined by Richard Hobbs in the Museum's own archive shows that the Museum's Secretary, Edgar Forsdyke, proposed a reward equal to the full market value of the find, then assessed at roughly 20,000 pounds, before recommending a 90 per cent reduction because the finders had concealed the discovery for four years. The figure that resulted, 2,000 pounds, was split equally: two separate payment orders were sanctioned, one for Sydney Ford and one for Gordon Butcher, each for 1,000 pounds, sent to their solicitor on 16 August 1946.

Press coverage from 27 August 1946 onward, in the London News Chronicle, The Times, the Daily Telegraph and other papers, reported the equal 1,000-pound reward to each man. Both were quoted as satisfied. Butcher told a reporter, 'I didn't expect to get anything like as much.' Against an average working-class wage of about five pounds a week in 1946, a thousand pounds was several years' income - a genuinely large sum for both men, not a windfall for one and a pittance for the other.

The intrinsic silver value of the entire thirty-four-piece hoard, calculated by Plenderleith on 26 July 1946 at the standing bullion price of 44 pence per troy ounce, came to only 191 pounds two shillings and sixpence. Forsdyke made the same point to the press at the time: 'The intrinsic value [of the hoard] is about 190 pounds.' The gap between that figure and the roughly 20,000-pound valuation of the find as an antiquity is the clearest illustration of what the Museum was actually paying for: history and craftsmanship, not metal.

What is documented as genuinely contentious is not the split of the reward but the four years of concealment that preceded it, and the reduced reward that concealment cost both men jointly.

The find that was never relocated

Archaeologists were not satisfied that the story ended with the inquest. Tom Lethbridge, a Cambridge lecturer in Anglo-Saxon and Roman antiquities, and Major Gordon Fowler, an experienced Fenland archaeologist, both gave expert testimony at the inquest and returned to the field afterward hoping to locate the exact findspot and search for any part of the hoard that might remain in the ground. Ford supplied a sketch map at the inquest, then a different account of the spot's location on at least two later occasions; when Lethbridge and Fowler dug at each location he indicated, they found no sign that a hole had ever been disturbed there.

The failure to relocate the findspot, despite a proper scheduling of the site by the Ministry of Works and a further organised excavation once the 1946 crop was cleared, fed real suspicion among archaeologists that Ford's account was incomplete or false. Lethbridge and Fowler came to suspect a connection to an unrelated, unresolved local rumour from the early 1920s about treasure supposedly buried decades earlier by a farmer known locally as 'Black Jack' Seaber. Richard Hobbs's 2008 examination of the British Museum's internal correspondence traced this theory in detail and found no evidence connecting it to the Mildenhall find; he concluded the likelier explanation is an unrelated, older cache of coins on the same land, if anything was buried there at all. A separate theory that the treasure was flown into the wartime airbase at Mildenhall from North Africa or Italy and buried near the runway is contradicted by the plain fact that Butcher's plough struck the hoard well away from the airfield, on Rolfe's farmland.

No part of the hoard beyond the thirty-four registered pieces has ever surfaced, and the precise spot where Butcher's plough struck it has never been rediscovered.

Roald Dahl's version, and why it is not this record

The writer Roald Dahl learned of the discovery from a newspaper report in April 1946, drove from his home in Great Missenden to interview Gordon Butcher, and sold an account of the story to the American magazine the Saturday Evening Post, which published it on 20 September 1947 under the title 'He Plowed Up $1,000,000'. Dahl revised the piece and republished it as 'The Mildenhall Treasure' in the 1977 collection The Wonderful Story of Henry Sugar and Six More, and it was issued again in 1999 as a standalone illustrated book by Jonathan Cape, illustrated by Ralph Steadman. This record found no evidence of a BBC television dramatisation of the story and none is claimed here; if one exists it was not located in the sources consulted.

Dahl built the piece as a deliberate literary contrast, casting Butcher as guileless and honest - 'a good, sharp, lean face, which was without a trace of malice or cunning or greed' - against a Ford drawn as calculating - 'middle-aged or a little older, baldheaded, long-nosed, with a clever, foxy look about his face.' Richard Hobbs's 1997 study in the journal Antiquity, examining Dahl's account against the documentary record, found that none of the specific claims Butcher made to Dahl were contradicted by other testimony, but that Dahl's characterisation of the two men was a conscious authorial choice rather than reporting. Dahl said as much himself, writing to a correspondent after the 1977 republication: 'the fact that my story may not be at all accurate doesn't worry me in the least. I am a fiction writer anyway and my job is to entertain... This is about the only non-fiction story I have ever written, so it is hardly surprising it turned out to be partly fiction.'

Dahl's story is culturally significant and is recorded here as reception history. It is not a source for any factual claim in this record about the discovery, the inquest or the reward; those rest on the coroner's inquest record and the Museum's own archived correspondence as reconstructed by Hobbs.

Dating and the question of manufacture

The dish's date rests on stylistic comparison rather than an inscription or a dated hoard context. When Christopher Hawkes of the British Museum first examined photographs of the hoard in June 1946, he initially proposed a third-century date and speculated that the hoard was buried during the Allectus usurpation of the 290s. He revised this after seeing the full assemblage, and the chi-rho monogram found on some of the hoard's spoons, a Christian symbol that could not have been used before Constantine's toleration of Christianity in 313, ruled out any date before the fourth century. The consensus since has been that the material principally dates to the fourth century CE, matching the British Museum's own current dating of the Great Dish to the 300s.

Where in the empire the dish was made is not settled. The British Museum's own published position is deliberately cautious: 'We cannot yet say where objects such as the Great Dish were manufactured, but it seems safe to surmise that it would have been somewhere in the general Mediterranean region.' Other scholarship has proposed the eastern Mediterranean specifically, on stylistic grounds, or Gaul, noting that at least one other piece in the same hoard, a flanged bowl with a niello-decorated cover, is separately catalogued by the Museum as made in Gaul. Neither the eastern Mediterranean nor Gaul is established for the Great Dish itself; both remain arguments, not documented findings, and this record treats them as such.

The comparison most often drawn is with the Seuso Treasure, a larger hoard of late Roman silver of disputed find circumstances now partly in Hungary's national collection, whose Dionysus ewer shares a high degree of stylistic similarity with the Bacchic frieze on the Mildenhall dish. The similarity supports a shared artistic environment across the fourth-century Roman world, without pinning down a single workshop for either object.

Why it was buried, and what is inferred versus known

The standard reading of the Mildenhall hoard, repeated in most popular accounts, is that it was buried for safekeeping amid the instability of the last decades of Roman rule in Britain and never recovered by its owner. The Roman garrison began its final withdrawal from Britain in the early fifth century, and burial dates for the hoard in the range of roughly 380 to 420 to 430 have been proposed on stylistic grounds by specialists working on comparable late Roman silver hoards, though this range is an estimate rather than a fixed date drawn from an inscription or coin.

The 'intention to recover' reading is also the legal foundation of the find's treasure trove status: at the 1946 inquest, both Lethbridge and Fowler argued as expert witnesses that the assemblage represented a hoard buried for later retrieval, not grave goods or a votive deposit, a distinction that mattered because English treasure trove law at the time required proof of an intention to recover for the Crown to claim ownership. They contrasted this explicitly with the Sutton Hoo ship burial of 1939, discovered only seven years earlier, which was not declared treasure trove precisely because its contents were interred as grave goods, never meant to be dug up again.

Beyond that legal argument, the causal link some archaeologists drew at the time between a specific documented period of unrest, such as the Allectus usurpation of 296, and the burial has since been abandoned along with the third-century dating that produced it. Hobbs, writing in 2008, is explicit that the reasoning connecting hoard burial to any one crisis was 'natural' for archaeologists working in 1946, shaped in part by the recent memory of a Nazi invasion threat, but that 'today it is recognised that a more complex range of interpretations might be invoked.' No documented evidence - a coin, an inscription, a dated destruction layer at the findspot - fixes either the exact burial date or the specific crisis, if any single crisis, that prompted it. This record treats the burial-for-safety reading as the argued consensus explanation, not a documented fact.

The local landscape

The find took place at the edge of the Fens, a low-lying wetland margin that was more agriculturally productive in the Roman period than its later reputation as marginal land suggests. Two long-distance Roman-period routes, the Icknield Way and Peddars Way, crossed the wider area and connected it to the rest of Roman Britain. In 1932, ten years before the treasure surfaced, Lethbridge and a collaborator named Leaf had excavated the remains of a small Roman building on the same landholding, about 27 to 55 metres from the spot Ford eventually indicated as the findspot, depending on which of his several conflicting accounts is used. The building had underfloor heating (a hypocaust), brick piers and a stokehole, but consisted of only two small rooms and an outbuilding, closer to a substantial cottage or a farm building than a villa in the grand sense; painted wall plaster and a scatter of late Roman pottery were the only other finds.

A separate and later excavation at Beck Row, in the same general area, uncovered the remains of a very large timber structure interpreted as a barn or grain store attached to a substantial farming estate. No excavation has established a confirmed structural connection between the Great Dish's owner and either building; both are offered in the literature as evidence for a genuinely prosperous late Roman presence in the immediate area, capable in principle of having owned an object of this quality, without proving that either specific building was that owner's residence.

Who might have owned it

No inscription on the Great Dish itself names an owner. Elsewhere in the hoard, two of the smaller Bacchic platters carry the scratched inscription 'eutheriou', read as Greek for '(the property) of Eutherios'. Kenneth Painter, author of the British Museum's first full catalogue of the treasure in 1977, proposed that the hoard, or at least this inscribed pair, may have belonged to a Roman general named Lupicinus, sent to Britain from Gaul in 360 CE by the emperor Julian to put down barbarian incursions, on the basis that a senior official named Eutherios is recorded in Julian's circle. Painter's own presentation of the theory treats it as unproved, and it has an obvious weakness: Lupicinus was recalled to Rome, fell from imperial favour, and was arrested there, meaning that if the silver was his, he was never in a position to return to Britain and recover it, which is at least consistent with the hoard never being reclaimed.

A second, separate line of argument rests on the chi-rho Christian monogram found on some of the hoard's spoons, taken by some scholars as evidence that the owner, or at least part of the household, was Christian - plausible for a person of rank in fourth-century Britain after Constantine, but not something the Great Dish's own purely classical, non-Christian iconography confirms or denies. Both theories concern the hoard's ownership in general. Neither is a documented fact about the Great Dish specifically, and this record grades both as argued rather than established.