I · THE OBJECT · BRITISH MUSEUM
The Lewis Chessmen
▶ Listen · narrated
Medieval chess sets almost never survive complete. This hoard came out of a Scottish beach with four sets' worth of pieces — and expressions no carver bothered with again for centuries.
At a glance
- Date
- 13th century
- Material
- Walrus ivory, with some pieces in whale tooth
- Discovery
- 1833, Isle of Lewis, Outer Hebrides
- Pieces found
- 93 gaming pieces plus one belt buckle
- Current location
- British Museum (82 pieces); National Museum of Scotland (11 pieces)
Look closer
The berserkers bite their shields
The warders — equivalent to rooks in the modern game — are carved as standing warriors. Most hold a sword and shield calmly. But eight of them have their shields raised to their mouths, teeth sunk into the upper rim, faces contorted. These are berserkers, the Norse warriors who fought in a trance state. The shield-biting pose appears in saga literature and on other Scandinavian carvings from the period. Whether it describes an actual battle practice or a symbolic gesture is disputed, but the carver clearly expected the reference to be recognised.
The queens rest a hand on one cheek
Twelve queens survive in the hoard. Each sits on a throne, and each raises her right hand to her face, fingers splayed against her cheek. The gesture reads as pensiveness, perhaps melancholy. It is not unique to these pieces — the pose appears in other Romanesque sculpture — but it is unusual in a chess set, where most medieval carvers gave royalty static, symmetrical authority. These queens look as though they are weighing a decision, or enduring something. The repetition of the gesture across all twelve suggests it was deliberate, not decorative accident.
The knights' horses look worried
Eight knights survive, each mounted. The riders sit upright, composed, but the horses' heads are turned sharply to the side, ears back, mouths open, eyes wide. They look startled or unhappy. Medieval carvers often gave animals more expression than human figures, but these ponies are unusually specific in their unease. Whether the carver was working from observation of real horses or simply enjoyed the contrast between calm rider and anxious mount is not recorded, but the effect is consistent across the set.
The story
The hoard was found in 1833, buried in a sandbank at Uig on the Isle of Lewis in the Outer Hebrides. Accounts of the discovery vary. Some say a cow uncovered the pieces by scraping at the sand; others that a local man found them in a stone chamber. What is consistent is that the pieces were together, hidden deliberately, and had been there for centuries.
Ninety-three gaming pieces came out of the sand, along with a belt buckle. Seventy-eight were chess pieces: eight kings, eight queens, sixteen bishops, fifteen knights, twelve warders, and nineteen pawns. The remainder were tablemen, used for a different game, possibly an early form of backgammon. The pieces are carved from walrus ivory, with a few made from whale tooth. All are Norse work, dated to the thirteenth century on stylistic grounds. The carving is confident, the details specific, and the scale consistent, which suggests a professional workshop rather than a single amateur carver working alone.
The pieces were not all carved by the same hand. Variations in style and finish indicate at least four or five different craftsmen contributed to the hoard. Some pieces are fully painted — red for one side, white or uncoloured for the other — while others show only traces of pigment. The paint that survives is bright: vermillion red, and details picked out in other colours now mostly lost.
No one knows why the pieces were buried on Lewis, or who owned them. The island was part of the Norse world in the thirteenth century, ruled by Norwegian kings until 1266, so their presence there is not in itself strange. What is strange is the quantity. Four nearly complete sets, plus extras, represent significant wealth. Chess was a high-status game, and walrus ivory was an expensive material, harvested in the Arctic and traded across long distances. Whoever buried this hoard either owned a trading stock or possessed a collection valuable enough to hide in a time of danger.
The pieces were sold shortly after discovery. Most went to the British Museum in 1831; eleven were acquired by the National Museum of Scotland in 1888. In 2019, a warder that had been in a family's possession since the nineteenth century, stored in a drawer and mistaken for a decorative figure, was identified and sold at auction. Four major pieces and many pawns remain unaccounted for, presumably still buried, lost, or unrecognised in private hands.
The style is Romanesque, consistent with Norwegian workshops of the period. The thrones, the drapery, the architectural details on the bishops' mitres — all point to Trondheim or another Norwegian centre with access to walrus ivory and to European artistic currents. The expressions, though, are not standard Romanesque. Most medieval chess pieces are static, hieratic, faces blank or symmetrical. These are not. The queens look thoughtful. The warders look wild. The bishops look variously bored, stern, or faintly appalled. Individuality of expression was not a priority in most medieval carving, which makes the Lewis pieces unusual, though whether that reflects the carver's skill, the patron's instruction, or simply a regional style is unclear.
Why it mattered then
Chess reached Europe from the Islamic world in the tenth or eleventh century, and by the thirteenth it was established among the Norse elite as both a game and a status marker. Saga literature mentions chess as a pastime of kings and heroes. Owning a fine set signalled wealth, taste, and participation in a cosmopolitan culture that stretched from Scandinavia to the Mediterranean. The Lewis pieces were luxury goods. Walrus ivory came from the Arctic — Greenland, Iceland, or the Norwegian Arctic coast — and was traded south through networks controlled by Norse merchants. Carving it required skill and time. The pieces in this hoard were not made for casual use. They were made for display, for gift-giving, perhaps for trade. A merchant travelling with several sets could sell them to wealthy patrons, or a lord could commission them as diplomatic gifts. The fact that they were buried suggests danger. Hoards are usually hidden in haste, with the intention of retrieval. Whoever buried the Lewis chessmen either never returned or never found them again. The thirteenth century was a period of political instability in the Hebrides, caught between Norwegian and Scottish claims. Burying valuables was a rational response to uncertain times.
Why it matters now
The Lewis chessmen are among the few surviving medieval chess sets, and the only large hoard of gaming pieces from the Norse world. Most medieval ivory was later melted down, reworked, or lost. These survived because they were buried and forgotten. They matter now because they show what medieval carvers could do when they chose to give figures personality. The berserkers, the thoughtful queens, the worried horses — none of these details were necessary. A chess piece only needs to be recognisable. But someone decided that a warder should look wild, that a queen should look pensive, that a horse should look startled. Those decisions, repeated across dozens of pieces, suggest a carver who was interested in expression, not just function. They also matter because they are legible. Modern viewers, who know nothing of Romanesque sculpture or Norse sagas, still read the queens as thoughtful and the berserkers as wild. The expressions work across eight centuries, which is rare. Most medieval art requires context to interpret. These pieces do not. A face biting a shield is a face biting a shield, and it looks as strange now as it must have looked then.
The surprising detail
One piece was kept in a family drawer for generations, mistaken for a decorative antique. It was inherited, stored with other small objects, and never examined closely. In 2019, the family brought it to an auction house, where it was identified as a Lewis warder. It sold for £735,000. The family had no idea what it was. The piece had been in their possession since the nineteenth century, acquired by an ancestor who likely bought it as a curiosity without knowing its provenance. It is now the most expensive chess piece ever sold at auction, and its rediscovery suggests that other pieces from the hoard may still be out there, unrecognised.
What is disputed
The circumstances of the discovery are not reliably documented. The pieces were found in 1833, but contemporary accounts differ on the details. Some sources say they were uncovered by a cow; others that they were found in a small stone chamber or cist. The name of the finder is also uncertain. What is clear is that the pieces were found together, buried deliberately, at Uig on Lewis. The original owner, the reason for burial, and the workshop that produced them remain unknown. Stylistic analysis points to Norway, probably Trondheim, but no documentary evidence ties the pieces to a specific carver or patron.
How it got here
The pieces were sold shortly after discovery in 1833. The majority were acquired by the British Museum; eleven pieces were purchased by the National Museum of Scotland in 1888. The pieces were found in Scotland, in a region that was under Norwegian rule in the thirteenth century but had been part of Scotland since 1266. There is no active repatriation claim from Norway, and the question of where the pieces 'belong' is complicated by the fact that they were Norse-made, found in Scotland, and have been in museum collections for nearly two centuries. Both the British Museum and the National Museum of Scotland present them as part of Scotland's medieval heritage, while acknowledging their Norse origin.
Remember this
Ninety-three pieces, carved with more expression than function required, buried and lost for six hundred years — and at least four are still missing.
Test yourself
The Lewis chessmen include queens, but many earlier Islamic and European chess sets do not. What changed in the game between the eleventh and thirteenth centuries that made the queen piece standard?
The piece that became the queen was originally a counsellor or vizier in the Islamic version of the game, moving only one square diagonally. As chess spread into Christian Europe, the piece was reinterpreted. By the thirteenth century, it was commonly understood as a queen, though its movement had not yet changed — that transformation, giving the queen its powerful modern move, came later, in the fifteenth century. The Lewis pieces show the queen as an established figure in the Norse game, seated on a throne and given individual expression, but still a century or two before she became the strongest piece on the board. The shift from vizier to queen reflects the game's localisation: each culture adapted the pieces to its own political and symbolic structures, and medieval Europe had queens, so the game acquired one too.
Go deeper
Image: National Museums Scotland, CC BY-SA 4.0. Licence: CC BY-SA 4.0. Source.