Pick up a backgammon set and you are touching one of the oldest games still played anywhere on earth. The 24 points and 30 checkers describe a structure that has barely changed in two thousand years, and parts of it are older than the pyramids. Mesopotamian kings played a version. Roman emperors played another. A Persian vizier reportedly built one as a riddle for the Indian court. And in a coffee shop in Istanbul or Athens this afternoon, two men will play a hand of it for nothing more than the next round of coffee.
Where It All Began
Most surveys of board game history trace the line back to two ancient ancestors. The first is the Royal Game of Ur, played in Mesopotamia around 2600 BCE, two boards of which were uncovered in the royal tombs at Ur in the 1920s by Sir Leonard Woolley. The board is not the same as a modern backgammon board, but the structural idea, a race of game pieces around a track governed by dice, is already present in recognizable form. The rules were a mystery for most of the twentieth century. They were reconstructed only in 1985, when Irving Finkel at the British Museum identified a cuneiform tablet that turned out to be a Babylonian rulebook.
The other ancestor commonly cited is Senet, the Egyptian board game found in tombs from the third millennium BCE. Senet shows up in wall paintings and in burial goods, and is named outright in the Book of the Dead. Players raced pieces along a track of thirty squares while reading the casts of throwing sticks. Whether Senet is a direct ancestor of backgammon or a separate branch of the same idea is still debated. What is not in doubt is that for at least five thousand years, humans have been racing tokens around boards by rolling something at them.
The line that leads straight into backgammon as we know it runs through Persia. Around the sixth century CE, a game called nard appeared in the Sasanian Empire. A Persian text from the period claims that nard was invented by a vizier named Burzoe as a riposte to chess, which had recently arrived from India. The story goes that Burzoe presented the game to the Indian court as a challenge, with the implication that any sage who could not work out the rules unaided would have to accept Persian superiority. Whether the story is true or constructed after the fact, what matters is that nard already had the essential structure of modern backgammon. A board of 24 points split into four sections, with 30 stones divided between two players and a pair of dice deciding every move.
From the Roman Carriage to the European Tavern
Rome had its own version, called tabula or alea, the latter just meaning “dice”. Tabula was widespread enough that Suetonius, writing in the early second century, claimed that the emperor Claudius wrote a treatise on it. None of the treatise survives. Suetonius also mentions that Claudius had a board built into his carriage so he could play while travelling, which would make him the first recorded long-distance tabula commuter in history. The game persisted through the Byzantine period in essentially the same form, and from there it crossed back into Europe by way of the Mediterranean trade routes through Italy and Spain.
Through the Middle Ages, the game travelled under many names. In France it was known as jacquet or tric-trac. In Germany as puff. In England, where the version eventually settled into something close to its modern shape, it was called “tables” through most of the medieval period. The name backgammon, in its modern English form, first appears in print in 1635. The etymology is disputed, but the most accepted theory traces it to a compound of “back” and the Middle English “gammen” meaning game, referring to the rule that allows captured pieces to be re-entered at the back of the board.
What is striking about this stretch of history is how little the rules changed. A player from sixth century Persia, dropped into a seventeenth century English tavern, would have recognized the board within a few minutes and the play within a few moves. Where chess transformed under Persian and European hands into something almost unrecognizable from its early forms, backgammon kept its bones.
The Cube That Changed Everything
Until the 1920s, backgammon was largely a parlor game. It was widely played but rarely studied. The single change that took the game from a casual pastime into a serious gambling pursuit was the introduction of the doubling cube. The cube is a six-sided die marked with the numbers 2, 4, 8, 16, 32, and 64. Either player can offer to double the stakes at any point in the game by turning the cube and offering it to the opponent, who must either accept the new stakes or resign at the current ones. The cube can then be doubled and re-doubled multiple times within a single game.
The doubling cube was introduced in New York around 1925. The exact moment and the exact inventor are not recorded, but the consensus is that it emerged in the city’s gambling clubs as a tool for raising stakes mid-game. What it did was transform backgammon from a race driven by luck into a contest of decision points driven by skill. Knowing when to offer the cube and when to accept it became as important as knowing how to play the pieces. Strong players began to out-earn weak ones by margins that had not been possible before. The change was so significant that the world before the cube and the world after it are sometimes discussed in the literature as different games entirely.
When Backgammon Went to Tournaments
Until the 1960s, backgammon was a private affair, played in clubs and at the table after dinner. The man who hauled it into public competition was Prince Alexis Obolensky, an exiled Russian aristocrat living in the United States. In 1964 he organized what is usually credited as the first international backgammon championship, held at the Lucayan Beach Hotel in the Bahamas. The Las Vegas tournament that became the World Backgammon Championship followed in 1967. Obolensky’s events drew an audience of Hollywood, Wall Street, exiled European nobility, and a handful of mathematicians who had worked out that the game was suddenly worth studying.
The Monte Carlo World Backgammon Championship was established in 1979 and has been held in some form most years since, making it the longest running major event on the calendar. The Nordic Open, founded later, draws large fields from across Northern Europe each year. Smaller national circuits exist across Europe and into Asia. Cities with strong backgammon traditions tend to anchor those circuits, and Athens, Istanbul, Tel Aviv, and Tokyo all host their own annual events.
At the highest level, backgammon stops being a race and becomes a duel of decision quality conducted under continuous dice noise. The hardest match to win in any of these tournaments is the final, and the hardest part of any final is staying composed across a sequence of cube decisions that may decide the match in two or three rolls.
The Game That Outlived Its Empires
What is unusual about backgammon is not that it has survived. Many games have. What is unusual is the route by which it has survived. The Royal Game of Ur, played in Mesopotamian palaces, vanished from Mesopotamia long before Irving Finkel reconstructed its rules at the British Museum. He discovered, in the course of his research, that a recognizable descendant of the game was still being played by Kerala Jews and by some communities in southern India. The ancient board game had crossed the world and survived in places where nobody at the British Museum had thought to look.
Backgammon’s line is the opposite. It has been continuously played in roughly the same form, on roughly the same board, for at least fifteen centuries. The Turkish tavla, the Greek tavli, the Lebanese sheshbesh, and the Iranian takhteh nard are not separate descendants. They are the same game with different local rules and different vocabularies for the score.
The other modern shift came from computing. In 1992, Gerald Tesauro at IBM published TD-Gammon, a neural network trained by playing against itself. Within a few years it was holding its own against the best human players in the world. By the late 1990s, programs like Jellyfish and Snowie, followed by eXtreme Gammon, had moved beyond holding their own and into territory that human players could study but not beat. The modern professional studies backgammon largely the way modern chess players study chess. Positions get fed to a strong program, and the work becomes understanding why the bot’s recommendation differs from the human instinct.
What that change has done to the live tournament is interesting. Players still gather in person and still roll dice between cups of coffee. But the level of preparation that separates a top player from a strong amateur is now largely a matter of how many hours each side has spent in front of XG.
Next time you sit down to a game and roll the first cast of the dice, take a moment to consider that the gesture is older than the alphabet and just barely younger than the wheel.
