Mancala, The National Game of Africa

By Stewart Culin

Director of the Museum of Archaeology and Paleontology, University of Pennsylvania.
Read before the Oriental Club of Philadelphia, May 20, 1894.
Published in the Report of the National Museum, 1894, Pages 597-611.

Note: For the most part, this is a direct transcription of the paper by Culin. His spelling and sentence structure have been maintained, but some punctuation has been altered. Words in Syrian or African languages with diacritical marks are transcribed from his text, and are presented within the limitations imposed by HTML. The paper was scanned from a photocopy of the printed text, edited in a word processor, then created as a Webpage. The graphics are slightly edited copies from sketches and photographs accompanying the original text.

Turkish Girls Playing Mancala

The comparative study of games is one that promises an important contribution to the history of culture. The questions involved in their diffusion over the earth are among the vital ones that confound the ethnologist. Their origins are lost in the unwritten history of the childhood of man. Mancala is a game that is remarkable for its peculiar distribution, which seems to mark the limits of Arab culture, and which has just penetrated our own continent after having served for ages to divert the inhabitants of nearly half the inhabited area of the globe.

The visitor to the little Syrian colony in Washington street in New York City will often find two men intent upon this game. They call it Mancala. The implements are a board with two rows of cup-shaped depressions and a handful or so of pebbles or shells, which they transfer from one hole to another with much rapidity. A lad from Damascus described to me the methods of play. There are two principal ways, which depend upon the manner in which the pieces are distributed at the commencement of the game. Two persons always engage, and ninety-eight cowrie shells (wada) or pebbles (hajdar) are used. One game is called La'b madjnuni, or the "Crazy game." The players seat themselves with the board placed lengthwise between them. One distributes the pieces in the fourteen holes, called bute, "houses," not less than two being placed in one hole.

Figure 1 - MANCALA - from a figure by Lane
Figure 1

This player then takes all the pieces from the hole at the right of his row, figure 1, G, called el ras, "the head", and drops them one at a time into the holes on the opposite side, commencing with a, b, c, and so on. If any remaining after he has put one in each of the holes on the opposite side, he continues around on [Page 598] his own row A, B, C.

When he has dropped his last piece he takes all the pieces in that hole and continues dropping them around as before. This is done until one of two things happen - his last piece drops into an empty hole, when he stops and his opponent plays, or it drops into a hole containing one or three pieces, completing two or four. In that case he takes the two or four pieces with those in the hole opposite, and if one or more of the holes that follow contains two or four without the intervention of a hole with any other number, he takes their contents with those opposite. The second player takes from the hole opposite and distributes his pieces around A, B, C. If the head is empty, the player takes from the next nearest hole in his row. When the board is cleared, each player counts the number he has above his opponent as his gains. No skill is necessary or of any avail in this game, the result being a mathematical certainty, according to the manner in which the pieces were distributed in the beginning. La'b hakimi, the "Rational game," or La'b akila, the "Intelligent game," is so called in contrast to the proceeding. Success in it depends largely upon the skill of the players. In this game it is customary in Syria to put seven pieces in each hole. The players, instead of first taking from the hole on their right, may select any hole on their side of the board as a starting place. They calculate the hole in which the last piece will fall, and the result depends largely upon this calculation. La'b rosëya is a variety of the first game and is played only by children. Seven cowries are placed in each hole, and the first player invariably wins. My [Page 599] Syrian friend told me that the, shells used in the game, are brought from the shores of the Red Sea. Mancala is a common game in Syrian cafes. Children frequently play the game in holes made in the ground when they have no board, a device also resorted to by travelers who meet by the way.

Figure 2 - Mancala Board - Jerusalem
Figure 2

A board in the Museum of Archaeology, University of Pennsylvania, from Jerusalem, is shown in figure 2, and one from Beirut, Syria, in plate 2, figure 1.

Plate 2, Figure 1 - Mancala Board
Beirut, Syria - USNM Cat. No. 164700
Plate 1

Mancala, the name which the Syrians give to this game, is a common Arabic word and means in this connection the "Game of transferring." It is not mentioned in the Koran by this name, but must have been known to the Arabs in the Middle Ages, as it is referred to ill the commentary to the Kitab al Aghani, the "Book of Songs," which speaks of a "game like Mancala."

Dr. Thomas Hyde gave a very good account of it two hundred years ago in his treatise, "De Ludis Orientalibus" (see figure 3: Mancala Board from a figure by Hyde), and Lane, in his "Manners and Customs of the Modern Egyptians," describes it very fully as played in Cairo upon a board with twelve holes, quite in the manner I have related. Seventy-two shells or pebbles are there used, and, [Page 600] whether shells or pebble, are indifferently called hasa.

Figure 3 - Mancala Board
Figure 3

The hemispherical holes in the board are called buyoot, plural of beyt. The score of the game is sixty, and when the successive gains of the player amount to that sum he has won. I soon found that I had learned from my Syrian acquaintance nothing that had not been recorded, but upon visiting the Damascus House in the Turkish Village at the Colombian Exposition at Chicago, I was enabled to engage with the Syrians in the game, and was impressed with the peculiar distribution of the game over the world. The Ceylon exhibit contained boards from the Maldives with sixteen holes in two parallel rows, with a large hole at either end. (figures 4 and 5.) Here the game is called Naranj.

Figure 4 - Board For Naranj (Mancala)
Maldive Islands - Cat. No. 16380,
Museum of Archaeology and Paleontology,
University of Pennsylvania
Figure 4

Figure 5 - Board For Naranj (Mancala)
Maldive Islands - Cat. No. 16379,
Museum of Archaeology and Paleontology,
University of Pennsylvania
Figure 5

Boards in the same exhibit from Ceylon had fourteen holes with two large center cavities (figure 6), the game being called Chanka. An Indian gentleman informed me that the game was common at Bombay.

Figure 6 - Board for Chanka (Mancala)
Ceylon - Cat. No. 16381
Museum of Archaeology and Paleontology,
University of Pennsylvania
Figure 6

His Highness the Sultan of Johore exhibited a boat-shaped board with sixteen holes (figure 7) under the name of Chongkak. I learned, too, that the game was common in Java, as well as in the Philippine Islands, where a boat-shape board with sixteen holes is also used (Plate 2, figure 2), [Page 601] the game being called Chungcajon.

Figure 7 - Board for Chongkak (Mancala)
Johore, Malay Peninsula - Cat. No. 16382
Museum of Archaeology and Paleontology,
University of Pennsylvania
Figure 7

It would thus appear that the game extends along the entire coast of Asia as far as the Philippine Islands. Mancala and a kind of draughts were the favorite amusements of the negroes from the French settlement of Benin on the west coast of Africa in the so-called Dahomey village at the Colombian Fair. They played on a boat-shaped board, with twelve holes in two rows, which they called adjito, with pebbles, adji, the game itself being called Madji.

Plate 2 - Figure 2 - Board for Chuncajon (Mancala)
Philippine Islands - Cat. No. 154195 - USNM
Collected by Alexander R. Webb
Plate 2

It is with the continent of Africa that the game of Mancala seems most closely identified. It may be regarded, so to speak, as the African national game. In the exhibit of the State of Liberia at Chicago, there were no less than eleven boards, comprising three different forms, said to be from the Deys, Veys, Pesseh, Gedibo, and Queah. (figures 8, 9, and 10.) They were catalogued under the name of Poo by which name the game is known to civilized Liberians.

Figure 8
Board for Poo (Mancala)
Liberian Exhibit
World's Columbian Exposition
Figure 8
Figure 9
Board for Poo (Mancala)
Liberian Exhibit
World's Columbian Exposition
Figure 9
Figure 10
Board for Poo (Mancala)
Liberian Exhibit
World's Columbian Exposition
Figure 10

The game is, in fact, distributed among the African tribes from the east to the west and from the north to the south. In Nubia, where a board with sixteen holes is used, it is known as Mungala.

Figure 11
Board for Gabattà (Mancala) - Abyssinia
From a figure in the
Sacred City of the Ethiopians by J.T. Bent
Figure 11

In the narrative of the Portuguese embassy of Alvarez to Abyssinia (1520-1527) reference is made to "Mancal" as an unknown game, antiquated in the reign of Don Manuel. Bent has recently described it as still existing in Abyssinia under the name of Gabattà2. (figure 11.)

Dr. [Page 602] George Schweinfurth states that it is played by the Niam-Niam, and is constantly played by all the people of the entire Gazelle district, although perhaps not known to the Monbuttoo. The Niam-Niam call the board, which has sixteen cavities, with two at the end for the reception of the cowrie shells, Abangah, (figure 12) and the Bongo name for the board is Toee.

Figure 12
Board for Abangah (Mancala)
Used by the Niam-Niam. From a figure in the
Artes Africanæ by George Schweinfurth
Figure 12
Plate 3
Board for Mbau (Mancala) - Elmina, Africa
Collected by W.H. Brown - USNM Cat. No. 151128
Plate 3
Figure 13 - Board for Kale (Mancala)
Gabon River, Africa
From a specimen in the Museum of
Archaeology and Paleontology, University of Pennsylvania
Figure 13
Plate 4 - Figure 1
Board for Kale (Mancala)
Falls of Gabon River, Africa
USNM Cat. No. 164869
Plate 4 Figure 1
Plate 4 - Figure 2
Óchi Board for Bau (Mancala)
Mount Kilimanjaro, Africa
Collected by W.L. Abbott
USNM Cat. No. 181805
Plate 4 Figure 2

He also says that it is found among the Peulhs, the Foolahs, the Toloofs, and the Mandingos in the Senegal countries, who devote a great portion of their time to this amusement. Rohlfs found it among the Kadje, between the Tsad and the Benne.3 It also occurs among the Biafren and the Kimbunda. Héli Chatelain, who lived from some time at Angola, described the game to me under the name Mbau, and said that cavities are cut in the rock for this game at the stations where the porters halt. A board collected by him at Elmina, is now in the U.S. National Museum, Washington, D.C., has twelve holes in two rows, with large holes at the ends. (Plate 3.)

Among the Fans of the Gaboon River the game is called Kale4 after the bean-like seed used in counting. (figure 13 and plate 4, figure 1.)

Another board in the U.S. National Museum, collected by that adventurous traveler, Dr. W.L. Abbot, from the Wa Chaga tribe at Mount [Page 603] Kilimanjaro, has twenty-six holes arranged in four rows of six each with two large holes at the ends. (Plate 4, figure 2.) He describes it in his catalogue, published by the Smithsonian Institution, under the name of Óchi, used for playing Bau, a common game throughout Africa, and says that it is played with nicker seeds and pebbles.

Bent, in his "Ruined Cities of Mashonaland," gives the following account of the game: "Huge trees sheltered the entrance to their village, beneath which men were playing Isafuba, the mysterious game of the Makalangas, with sixty holes, in rows in the ground. Ten men can play at this game, and it consists of removing bits of pottery or stone from one hole to another in an unaccountable manner. We watched it scores of times while in the country, and always gave it up as a bad job, deciding that it must be like draughts or chess learned by them from the former civilized race who dwelt here." He then proceeds to identify Isafuba with the games of Wari played on the west coast of Africa.

Prince Momolu Massaquoi, son of the King of the Vei tribe, described to me the manner of playing the game among the Vei. They call the game Kpo, a word having an explosive sound resembling a note of the xylophone, mimicking the noise made by the seeds or ivory balls with which the game is played when tossed into the holes on board. The boards, which are made with twelve holes in two rows, with large holes in the ends, are called by the same name. The boards used by the chiefs are often very expensive, being made of ivory and ornamented with gold. He had seen boards which cost 20 slaves. The holes in the boards are called kpo sing or kpo kungo, kungo meaning "cup." The game is usually played with sea beans which grow on vines like the potato on the west coast, or by the chiefs with the before-mentioned ivory balls. These seeds are called kpo kunje, kunje meaning "seed." He identified a board from the Gabon River as suitable for the game, although he said that much more elaborate ones, like those in the Liberian exhibit, were common. The depression in the middle of the board from the Gabon River is intended to catch pieces that do not fall in the hole for which they are intended. Cheating is practiced, and to guard against it players must raise their arms and throw the pieces upon the board with some violence. Two, three, or four play. The game differs somewhat from that played in Syria and Egypt. A player may commence at any hole on his side. His play [Page 604] ends when the pieces first taken up are played. He wins when the number in the last hole is increased to two or to three. He does not take those in the hole opposite. When two play, four beans are put in each hole, but when three or four play three beans are put in each hole. When two play the pieces are dropped around in the same direction as in the Syrian game, but when three or four play they may be dropped in either direction. When two play, each player takes one side of the board; when three play, each takes four holes, two on each side, dividing the board transversely into three parts, and when four play, each takes three holes. When two play, a winner takes only what he "kills" (); but when three or four play, when one completes two or three in a hole by his play, he takes those, in the next hole forward. When a man takes a piece, with one next to it, he uses his fingers to squeeze the pieces into his hand, the operation being called "squeezing" (boti), but this can only be done when one of the pieces is in one of the player's own cups and the other one or two in that of an opponent. Players sit cross-legged upon the ground, and when the chiefs play large numbers often assemble to watch them. I have given Prince Momolu's account somewhat at length, as several African travelers have declared the game incomprehensible to a white man.

Dr. Schweinfurth regards the Mohammedan Nubians as having received Mancala from their original home in Central Africa, and says that the recurrence of an object even trivial as this is an evidence, in a degree indirect and collateral, of the essential unity that underlies all African nations. Mr. Bent justly says that the game is found in some form or another wherever Arabian influence is felt, but, continuing, states that it forms for us another link in the chain of evidence connecting the Mashonaland ruins with an Arabian influence. Dr. Richard Andree, in his well-known work on Ethnological Parallels5 in which he has brought together many accounts of the game, says that be regards its progress from west to east, from Asia to the coast of the Atlantic. This opinion I share, Peterman relates that Mancala is played in Damascus with pebbles which pilgrims collect in a certain valley on their way from Mecca. From the comparatively early mention of the game in Arabic literature, and the retention of its Arabic name in Africa, Arabia would appear to be the source from which it was disseminated. Mohammed proscribed the Meiser game; and games of hazard, although played, are regarded by Mohammedans as prohibited by their religion. Mancala, a game of fate or calculation, appears to be looked upon with toleration, and it is not unreasonable to suppose that its wide diffusion is due to its having been carried by returning pilgrims to the various parts, of the Mohammedan world. If we accept this theory of its distribution. we have yet the more difficult question of its origin. This, I fear, is not to be determined directly, and will only be surely known when we attain a greater knowledge of the rules [Page 605] or laws which underlie the development of games, even as they do every other phase of the development of human culture.

Plate 5
Plate 5
Board for Wa-we (Mancala)
Island of St. Lucia
Collected by Rev. F. Gardiner
USNM Cat. No. 151286

I have recently been informed that Mancala is a common amusement of the negroes of San Domingo, who play upon boards scooped with holes. A board in the U.S. National Museum (Plate 5) was collected by the Rev. F. Gardiner, jr., in St. Lucia, where the game is played by the negroes under the name of Wa-Wee.6

It is not unlikely that Mancala may some day take its place among our own fireside amusements, when this account may answer some inquiries that may be made as to its history.

Since the above was written I have learned that the game of Mancala was published in the United States in 1891, under the name of Chuba, by the Milton Bradley Company, of Springfield, Mass., who furnish the following rules and account of the game:

Figure 14
Figure 14
Position of men at opening of the game

Chuba is an adaptation from a rude game of eastern Africa which is greatly enjoyed by the natives, who squat on the ground and play in holes scooped out of the sand, using shells, young cocoanuts, etc., for counters, which they move from hole to hole. As now presented to the civilized world for its diversion, Chuba is a game of skill for two players. It is made up of a board with 4 parallel rows of holes or pockets, 11 in each row, and 60 small beads used as men or counters. (See figure 14.)

The board is placed between the players as usual, with the longer sides next to them. Each one confines his playing to the two rows of pockets nearest him. The row close to his edge of the board is his outer row, while the other is his inner row. [Page 606]

Before beginning the game each player places a single counter in each of the pockets of his outer row and two counters in each pocket of his inner row, except that the pocket on his extreme left in the inner row is kept vacant and the one next to it holds but one man. The above diagram shows the arrangement of the board at the opening of the game. As indicated by the arrows, all moves in the inner row are from right to left, and those in the outer row from left to right. As the players face each other the moves in the two inner rows are necessarily in opposite directions.

The privilege of playing first in the first game is left to agreement or chance, not being regarded as of any consequence. In subsequent games the player who was victor in the last contest takes the lead.

The first player chooses any pocket in his inner row which contains more than one man from which to start his first move, and begins the same by picking up all the men in that pocket and dropping one of them in each of the consecutive pockets to his left until all the men in his hand have been distributed. If the last counter drops into a pocket that is occupied, the player continues the move by picking up all the men in that pocket, including the one dropped, and disposing of them as before. His move must continue in this same way until the last counter in his hand falls into an empty pocket, and the move may extend around the course, into the outer row, or even farther, as indicted by the arrows.

Figure 15
Figure 15

If this empty pocket into which the last man falls is in the inner row and has opposite it a pocket in the opponent's inner row containing one or more men, the player captures these men and at once removes them from the board. And if there are one or more men in the corresponding opposite pocket of the opponents outer row, they must also be taken. Furthermore, he must select another pair of opposite pockets in his opponent's rows from which to remove any men that they contain. In making this choice he is at liberty to pick out any pair of opposites, whether both are occupied or empty, or one is occupied and the other empty. The accompanying diagram will explain the meaning of this rule. (See figure 15.)

Suppose the player B had just finished a move by dropping a "last man" in No. 1. He can capture all the men in 2 and 3 by his skill and also in 4 and 5 or from any other two opposite pockets of his opponent's inner and outer rows. Had 2 been vacant, however, he could not have taken any men. Had 3 been vacant, he could have taken the men from 2 and those from 4 and 5. Had his last man fallen in the outer row, in 6 for example, the effect would have been of no avail in capturing anything, because the outer row is always noncombatant.

A man in the outer row can not be moved until he has been played upon by a man from the inner row.

A move can not begin from a pocket holding a single man if the player has a pocket containing more than one man. When a move does begin from a pocket containing a single man, it can not be played into an occupied pocket. [Page 607]

When all the men which a player has become single, those remaining in his outer row which have not been played on are forfeited to the opponent.

The winner is the player who captures all his opponent's men.

It is an advantage to a player to get his counters singled as soon as possible, unless he sees that his opponent is doing the same thing, when a different policy is wise.

If he wishes to cover two or three vacant spaces in order to effect a capture, it can often be done, provided he begins his move far enough back from those vacant pockets.

The loss of counters during the earlier part of the game is not necessarily as great a disadvantage as in most games, because so much depends on the final move, in which there is the chance for a brilliant display of skill.

The native players of the original Chuba say "chee" at the end of each move which gives notice to the opponent to proceed; and toward the close of the game, when the moves follow in rapid succession, the effect is very amusing.

The natives call the counters in the inner row "man and wife," and those in the outer row "spinsters." But these spinsters are married by passing a counter over them from the inner row, till, in the progress of the game, all the pieces become single, when they are all called "widows." These widows have a double advantage over the married families, and are sure to make havoc among them. The game is appropriately named, as the word chuba means "to extinguish" or "eat up," and the object of each player is to annihilate his opponent by putting the latter's counters in a position from which escape is impossible.


  1. Speaking of the peasants of Sallaba, he says. "These primitive people are perfect artists in cow dung. With this material they make big jars in which to keep their grain, drinking goblets, and boards for the universal game, which the better class make of wood. I brought one of these away with me to show how universal this game is among the Abyssinians, from the chief to the peasant, and it reached the British Museum unbroken. This game is called Gabattą, and the wooden boards made by the better class contain eighteen holes, nine for each person. There are three balls, called chachtma, for each hole, and the game is played by a series of passing, which seemed to us very intricate, and which we could not learn; the holes they call their toukouls, or huts, and they get very excited over it. It closely resembles the game we saw played by the negroes in Mashonaland, and is generally found in one form or another in the countries where Arab influence has at one time or another been felt." (The Sacred City of the Ethiopians, London, 1873, pp. 72-73.)
  2. Richard Andree, "Ethnographische Parallelen," neue folge, Leipzig, 1889, P. 102.
  3. The collector, Rev. A.C. Good, gives the following account of the game: "Two players seat themselves on opposite sides of the board, and four counters are placed in each of the twelve pockets. Then one player takes the counters out of a pocket on his own side and drops one in each pocket around as far as they will go, going to right and back on his opponent's side in the opposite direction from that in which the hands of the clock move. They move thus alternately until one manages to make his last counter fall in a pocket on his opponent's side, where there were only one or two counters. When he had done so he had won the counters in that pocket, including his own last counter. These he transfers to the receptacle in the end of the board to his right. A single counter taken from last pocket on player's right can not win from opponent's fast pocket opposite, even though it contains only one or two counters. When a pocket has accumulated twelve or more counters, so that a player drops clear around and back to where he began, he must skip the pocket from which be started. When so few counters remain in the pocket on the board that no more can be won, the game is ended and each counts his winnings. The counters that remain in the board at the end of the game are not counted by either player. The game is sometimes varied thus. When a counter wins as above, not only the contents of that pocket is won, but of the pocket or pockets before it on the opponent's side that has contained only one or two counters back until one is reached that has been empty or had three or more counters before the play. This last is rather the better game of the two. The Fans do not play these games skillfully. They seem unable to count ahead to see where the last count will fall. A white man, as soon as he understands the game, will beat them every time.
  4. "Ethnographische Parallen," neue folge, Leipzig, 1889, p. 101.
  5. Mr. Gardiner writes in a letter to Dr. Brown Goode under date of May 2, 1895: "The game of Wa-wee was bought in St. Lucia, but I found it in use also in Barbados and Martinique among the negroes. As far as I could ascertain, they supposed it very old - came from their fathers. I supposed it came from Africa; but no one seemed to know anything about it. It is a regular gambling game." In regard to the method of play, he says: "As near as I can remember, each of the small side holes has a given number of beans put in, each man taking one side and one large hole as a goal. The beans are taken up from one hole in the hand and dropped in a certain order in the other holes, going round the whole circle. If the last one drops in a hole which has a certain number of beans in it (I don't remember the, number), he picks that lot up and goes on. The object is to land the most beans of your own and taken from your adversary in the end holes."

Last update January 10, 2010