Kalah

Rows of pits, long before the name

Games where you scoop up a handful of seeds and drop them one by one round a ring of pits are old and widespread. They turn up across Africa, along the Red Sea, through the Arabian peninsula and into South and Southeast Asia. Nobody needed a board: two rows of hollows in the ground and a pocketful of pebbles will do.

That simplicity is what makes the early history hard. A row of small round depressions in rock might be a game board, or a grinding surface, or a tally, and the hollows themselves cannot be dated. Confident claims about the oldest mancala board should be read with that in mind. These games are old, they spread early, and the details are thin.

Two-row games in Africa and the Middle East

Kalah descends from the two-row branch of that family: boards of six or seven pits a side, played across West Africa, the Horn, the Arabian peninsula and the Levant. Most of those games have no stores and capture on the count a pit reaches rather than on landing in an empty one. Kalah keeps the two rows and the sowing, then swaps in a store you sow into and a capture triggered by an empty pit on your own side.

The mid-century board that fixed these rules

This exact rule set was assembled in the United States around the middle of the twentieth century by a man who had met sowing games abroad and set about selling a neat version of them under the name Kalah. He patented the board and pushed it hard, especially at schools, on the grounds that it taught counting and planning.

That commercial push is why Kalah, out of hundreds of sowing games, is the one most people outside Africa and Asia have played. It is also why its rules barely drift: they were printed on a box rather than passed between players. The variants here differ only in how many seeds each pit starts with.

Solved by computer

Fixed rules and a small board made Kalah a natural target for game-solving research. In 2000 a group of computer scientists searched several Kalah boards exhaustively and showed that with six pits and four seeds, the first player can force a win. Larger boards, including the six-seed one, were settled later with deeper searches.

Solved does not mean the winning line is easy or memorable. It means a machine can prove the result from the opening position. No human plays perfectly, and the second player still has plenty of chances against another human.

What that means at the table

For a casual game it changes almost nothing: two players of similar strength will still have a good fight, and the fun is in spotting chains and traps a few turns ahead. For rated play it matters, because a fixed advantage to whoever moves first is not something a rating system should be measuring.

The usual fix, borrowed from other solved games, is the pie rule, and that is what this site uses. The opener plays, the second player decides whether to take that position over, and the built-in advantage is handed back to the players to argue over.

Fontes

• H. J. R. Murray, *A History of Board Games Other Than Chess* (1952) — the first broad attempt to classify the sowing games.
• Philip Townshend, *Mankala in Eastern and Southern Africa: A Distributional Analysis* (1979) — on where the different sowing games are actually played.
• Alex de Voogt, *Mancala Board Games* (1997) — a short, careful survey of the family and how people play it.
• Laurence Russ, *The Complete Mancala Games Book* (2000) — a wide collection of regional rule sets, including the commercial Kalah rules.
• Geoffrey Irving, Jeroen Donkers and Jos Uiterwijk, *Solving Kalah* (2000) — the computer search that settled the small Kalah boards.