Mancala
Mancala is a family of two-player board games in which players sow small pieces—seeds, stones, shells, or glass beads—around a board of pits or hollows, capturing opponents' pieces by counting and by the pattern of the sowing.
14 frameworks, from 1200 to today · Tabletop Games · Games
- Manqala 1200-Present
- Two-row mancala games 1700-Present
- Ayo 1800-Present
- Bao la kiswahili 1800-Present
- Congkak 1800-Present
- Four-row mancala games 1800-Present
- Oware 1800-Present
- Pallanguzhi 1800-Present
- South Asian pit-and-count games 1800-Present
- Southeast Asian mancala 1800-Present
- Tsoro 1800-Present
- Ethnographic documentation 1900-Present
- Kalah 1940-Present
- Combinatorial game analysis of mancala 1990-Present
Overview
Mancala is a family of two-player board games in which players sow small pieces—seeds, stones, shells, or glass beads—around a board of pits or hollows, capturing opponents' pieces by counting and by the pattern of the sowing1. The name comes from the Arabic naqala, "to move," and it is a convenience term rather than a single game: there is no one mancala, only a large family of related games played across Africa, Asia, and the Middle East, with dozens of distinct rule sets. What is at stake in the history is a puzzle of transmission and invention. Mancala games are among the oldest and most widely distributed board games in the world, yet they were played with ephemeral materials—hollows scooped in earth, seeds, pebbles—so they left almost no archaeological trace. Their history has to be reconstructed from travellers' accounts, ethnographic fieldwork, and the internal logic of the games themselves.
Sowing and capturing: the core mechanics
Nearly all mancala games share a board of two or more rows of pits, with a store at each end, and a rule that a player lifts all the pieces from one pit and drops them one by one into successive pits around the board. The distinctive feature is the sowing (also called sowing or broadcasting) and the capture: depending on the game, a player captures pieces when the last seed of a sowing lands in an empty pit, or in a pit opposite an occupied one, or when a pit reaches a certain count1. These rules are not arbitrary decoration; they are the game's strategic engine, and the differences between them define the major families.
Two broad families are usually distinguished. In the two-row games, each player owns one row of pits and a store; the classic example is Oware (also spelled Awari, Warri, or Wari), played in Ghana and across West Africa, in which a player captures when the last seed lands in an opponent's pit containing two or three seeds. In the four-row games, the board has four rows and play is more complex; the best-known is Bao la kiswahili, played on the Swahili coast of East Africa, especially in Tanzania and Kenya, with a large board and elaborate rules for moving seeds between rows.
Early evidence: from the Red Sea to the Sahara
The earliest written evidence for mancala-like games is often cited from the Kitab al-Aghani, a tenth-century Arabic collection of songs and poems, which mentions a game called qirqis or manqala played in the region of Mecca. The attribution is debated, and the passage is not a rule book. More secure are the rock-cut boards found across the Sahara, the Nile Valley, and the Red Sea hills: rows of small hollows pecked into stone, sometimes in the characteristic two-row or four-row pattern. These are difficult to date, but many are associated with caravan routes and with the spread of Islam, and they show that sowing games were established across a wide belt of Africa and western Asia well before the early modern period.
The stone boards are the closest thing mancala has to an archaeological record, and they are ambiguous: a row of hollows could have held seeds for a sowing game, counters for a different game, or nothing at all. What they do establish is that the board form—a line of pits—was familiar across a huge area, and that the games were played outdoors, in public, and with whatever materials were at hand.
Regional games and their names
The history of mancala is largely the history of its regional variants, each with its own name, board, and rules. In West Africa, Oware is the best documented: it was recorded by European traders and colonial officials from the eighteenth century onward, and it remains the region's most widely played mancala game. In Nigeria, the Yoruba play Ayo (also Ayoayo), a two-row game with its own capture rules, often on a carved wooden board with twelve pits and two stores. In Ghana, Oware and Warri are played on similar boards; the game is a social institution, played by men in public, and it has been used as a metaphor for strategy and patience in Akan culture.
In East Africa, Bao la kiswahili is the prestige game of the Swahili coast, played on a four-row board with a complex set of moves and captures; it is associated with the towns of Zanzibar, Dar es Salaam, and Mombasa, and it has a body of proverbs and etiquette attached to it. In Southeast Asia, Congkak is played in Malaysia, Indonesia, and the Philippines, usually by women and children, on a boat-shaped wooden board with two rows of pits and two large stores; the game is often decorated and is a common household object. In South Asia, Pallanguzhi is played in Tamil Nadu and Sri Lanka, on a board of fourteen pits, with rules for capturing by counting; it is one of several related games in the region12. In Zimbabwe, Tsoro is a four-row game played by the Shona, with a board of four rows of six or seven pits; it is one of the many regional names for games that share the sowing mechanic but differ in detail.
These names are not translations of one another. Ayo, Oware, Bao, Congkak, Pallanguzhi, and Tsoro are distinct games with distinct rules, and a player of one would not necessarily recognize the others as the same game. The family resemblance is real, but the differences are the substance of the tradition.
Ethnographic documentation and the colonial record
The systematic documentation of mancala games began in the nineteenth and early twentieth centuries, as European travellers, colonial administrators, and missionaries recorded the games they encountered. These accounts are uneven: some give detailed rules, others only describe the board. The value of the record is that it captures games at a particular moment, before widespread urbanization and the spread of manufactured boards changed play. The limitation is that the observers often misunderstood the rules or imposed their own categories.
In the twentieth century, anthropologists and ethnomusicologists began to study mancala games as social institutions rather than curiosities. They recorded not only rules but also the contexts of play: who played, where, when, and what the games meant. This work established that mancala games were often associated with particular social groups—men, elders, or children—and that they served as venues for gambling, courtship, and the display of skill. The documentation is still incomplete: many regional variants have never been fully described, and some are known only from a single account.
Kalah and the modern board game
The modern commercial history of mancala begins with Kalah, a two-row game patented in the United States in 1940 by William Julius Champion Jr. Champion's game was a simplified version of the sowing mechanic, with six pits per player and a store, and rules designed for mass production. It was sold as a wooden board with glass stones and became a popular family game in the United States and Europe. Kalah is not a traditional game; it is a modern invention that drew on the mancala family and standardized it for a commercial market3. Its importance is that it introduced millions of players to the sowing mechanic and made "mancala" a household word in the West, even as the traditional games continued to be played in their home regions.
Combinatorial game analysis
In the late twentieth century, mancala games attracted the attention of mathematicians and computer scientists working in combinatorial game theory—the study of games of perfect information with no chance, where the outcome can in principle be calculated. Oware and Kalah are both finite, deterministic games, and they are small enough to be analyzed by computer. The first strong programs for Oware were written in the 1990s, and they demonstrated that the game has a rich strategic depth despite its simple rules. The analysis has not "solved" Oware in the sense of a complete winning strategy, but it has clarified the value of particular positions and the importance of tempo and pit control. This work is part of a broader movement in which traditional games are studied with the tools of artificial intelligence, and it has fed back into how the games are taught and played.
The present
Today mancala games are played across their traditional range and beyond. Oware, Ayo, Bao, Congkak, Pallanguzhi, and Tsoro remain living traditions, taught by elders and played in schools and clubs. International tournaments exist for Oware and Bao, and digital versions of Kalah and Oware are widely available. The scholarly record is still being built: ethnographers continue to document regional variants, and computer scientists continue to analyze the games' strategic structure. The open questions are the old ones—how the games spread, how old they are, and how the many variants are related—and they remain difficult to answer because the evidence is thin and the games leave no ruins. What is clear is that the sowing mechanic is one of the most durable and adaptable inventions in the history of games, and that its many forms are still being played, taught, and studied.
References
- Mancala (introduction, General gameplay, Equipment, Capturing, Names and variants), Wikipedia
- Pallanguzhi, Wikipedia
- Kalah, Wikipedia
The frameworks of Mancala
Each framework is a way of working the field has used: when it held, and the question it answers.
- 1200-Present
- 1700-Present
- 1800-Present
- 1800-Present
- 1800-Present
- 1800-Present
Four-row mancala games
How do four-row sowing games differ from two-row ones?
- 1800-Present
- 1800-Present
- 1800-Present
South Asian pit-and-count games
What games are played by sowing and counting in South Asia?
- 1800-Present
- 1800-Present
- 1900-Present
Ethnographic documentation
How were mancala games recorded by outsiders?
- 1940-Present
- 1990-Present
Combinatorial game analysis of mancala
Can sowing games be solved by calculation?
How the ideas connect
Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.
Two-row mancala games grew out of Manqala
The Arabic term manqala and the sowing-and-capture mechanic it named were carried by traders and Islamic expansion into West Africa, where the two-row family (Oware, Ayo, Warri) developed as a distinct branch of the same game complex.
Four-row mancala games grew out of Manqala
Four-row mancala games such as Bao and Tsoro are a regional elaboration of the same sowing mechanic that the Arabic manqala named; the four-row board is a structural branch of the family rather than a separate invention.
Ayo grew out of Two-row mancala games
Ayo is a Yoruba two-row game whose rules and board belong to the two-row family; it is one of the regional variants through which the two-row tradition is documented.
Congkak grew out of Two-row mancala games
Congkak is a Southeast Asian two-row game with stores at each end, carried by Malay and Indonesian trade networks from the same two-row sowing tradition.
Oware grew out of Two-row mancala games
Oware is the best-documented member of the two-row family, with its capture-on-two-or-three rule and twelve-pit board defining the type.
Pallanguzhi grew out of Two-row mancala games
Pallanguzhi is a South Asian two-row counting game whose fourteen-pit board and capture-by-count rules place it within the two-row family.
South Asian pit-and-count games grew out of Two-row mancala games
South Asian pit-and-count games such as Pallanguzhi share the two-row sowing mechanic and counting captures, forming a regional branch of the two-row family.
Southeast Asian mancala grew out of Two-row mancala games
Southeast Asian mancala games, notably Congkak, are two-row games with stores, carried by Malay and Indonesian maritime trade from the wider two-row tradition.
Bao la kiswahili grew out of Four-row mancala games
Bao la kiswahili is the prestige four-row game of the Swahili coast, with elaborate rules for moving seeds between rows that mark it as a developed branch of the four-row tradition.
Tsoro grew out of Four-row mancala games
Tsoro is a Shona four-row game played on a board of four rows of six or seven pits, one of the regional names for the four-row sowing games.
Beyond the field
Where Mancala's ideas came from in other fields, and what they led to there.
Combinatorial game analysis of mancala influenced by Combinatorial game theory in Abstract Strategy Games
Combinatorial game theory, formalised by Berlekamp, Conway and Guy in Winning Ways (1982), supplied the game-tree and partizan-sum methods later applied to mancala variants. Analyses of mancala as a combinatorial game treat positions as sums of components in the CGT framework.
Carried by Winning Ways (1982); Elwyn Berlekamp; John Horton Conway.
Combinatorial game analysis of mancala influenced by Computer solving and engine analysis in Draughts
Techniques of computer solving and engine analysis developed for draughts and other abstract games, such as alpha-beta search and endgame databases, were carried over to mancala programs. The same retrograde-analysis and search methods used to solve checkers and draughts positions were applied to Kalah and Oware.

Mancala is a planet in Nousfarer
Its city has a tower for each of these 14 frameworks. Land, walk in, and climb one concept at a time.
Land on MancalaRelated fields
Fields its ideas are linked to
- Abstract Strategy Games 1 link
- Draughts 1 link