Continuing the series 'Ludi', and following on from ancient Egypt, here are selection of games played by ancient Greeks and Romans.
Pente grammai
Pente grammai, or ‘Five Lines’, was mentioned around 600 BC in a poem by Greek poet Alkaios. Gaming tables in terracotta with the typical five parallel lines have been found in graves dating to the same period. Game boards of the same type can sometimes be found scratched into the marble floors of temples and public buildings in Greek archaeological sites. As numerous vase paintings demonstrate, such as the one pictured above from the Vatican Museum, during the decades around 500 BC the Athenians imagined Ajax and Achilles playing the game during the siege of Troy. Five Lines, the ‘game of the heroes’, was regarded as a noble game for centuries and very popular.
It is only in the Imperial Roman period that we learn more about the game. In his Onomasticon, the 2nd-century AD lexicographer, Pollus, lists Pente grammai amongst games of chance. His entry states that ‘each of the two players had five pieces upon five lines’ adding that ‘there was a middle one called the sacred line. And moving a piece already arrived there gave rise to the proverb “he moves the piece from the sacred line” (bad luck!)’. The term 'sacred' may relate to the ancient Greek concepts of asylum and hiketeia, that is the inviolable right of persons in search of aid to take refuge in a sanctuary where nobody had the right to remove them by force. As such it describes well the function of this line. Much later, in the 12th-century AD, Eustathios explains that ‘the beaten player goes to it last’, that is the winning player is the one who manages to move all his five pieces onto the sacred line first. Based on these and other written and archaeological sources, a set of rules have been developed that are in keeping with these hints.
From archaeological finds it may be inferred that enlarged versions of the game also existed, notably a double version with 11 lines. Moreover, the existence of game boards with 2 x 5 and 2 x 11 squares in Roman cities such as Ephesus suggest that the Romans played the game within squares instead of upon lines as the Greeks did.
Gameplay: As far as the modes of playing Pente grammai are concerned, we may draw the following conclusions:
The standard game was played on aboard with five parallel lines. Larger versions could have more lines, but always an odd number as there had to be a central, or 'sacred', line.
From the written sources we learn that the aim of the game was to move all one's counters onto this 'sacred' line.
The number of counters used corresponded to the number of lines, each player having as many counters as lines on the board. The points, holes or circles at both ends of the lines on some of the boards suggest that one counter only was normally placed at the ends of the lines.
The opening position has one player's pieces at the end of each line on their side of the board, and the other player’s pieces are positioned opposite.
Each player takes it in turns to roll a die and moves their pieces anticlockwise around the board according to the number thrown. For example, if a player throws '6', then they can either move one piece six lines (or squares) or move several pieces to the total value of '6'. In either case, a piece can only end its move on an unoccupied line or square. Only the sacred line, or the two squares opposite each other in the centre of the board, can be occupied by more than one piece at a time, and both players' pieces can occupy their side of the sacred line simultaneously.
A player has to move if they are able to, but if they cannot, then their turn is lost.
The aim of the game is for each player to move their pieces so that all five eventually occupy the sacred line in the opposing player's half of the board.
Pentelitha
Pentelitha, where pente means ‘five’ and litha means ‘stones’ in ancient Greek, was probably played on the same boards as the Roman game Latrunculi (see below), with grids of either 12 x 8 squares or 10 x 11 squares. Of note, Pentelitha requires more stones (gaming pieces) than Latrunculi, hence the need for larger playing boards. The Roman version of the game appears to have been called ‘Calculi’.
Gameplay: The game is simple. The players start with equal numbers of different coloured stones, traditionally black or white. Each player takes turns to place one stone within a square on the board. The aim is to be the first to connect five stones in a row in a horizontal, vertical or diagonal line, hence the game’s alternate name of ‘Five-in-a-Row’. Clearly each player must block his opponent’s attempts to achieve a line of five stones.
The traditional rules of Pentelitha, or ‘Five-in-a-Row’, are thus:
Black plays first.
The first person to line up five stones in a row vertically, horizontally or diagonally wins.
Making a ‘double open-ended three’ is banned unless one is forced to do so.
If the board becomes full, the game is a draw.
A double open-ended three, or three in a row simultaneously in two directions, is banned because it is too easy to win, but can occur frequently. The addition of this rule makes for a much more interesting game and leads to the strategy in which one tries to make a double ‘three and a four’, which is like a double open-ended three, except that one line is made of four in a row.
Latrunculi
Latrunculi was the most popular ‘thinking man's game’ in the Roman Empire. The name comes from the word latrones originally meaning ‘mercenaries’ but which also can refer to ‘soldiers’ in the early Roman period. Later the term came to be used for highwaymen, or ‘robber-soldiers’ as they were known. Archaeological finds in Italy and Britain suggest Latrunculi had playing pieces of different coloured glass, sometimes precious stones, and of two types: the ordinary gaming pieces and an extra one variously called Rex (‘King’), Dux (‘Duke’) or Aquila (‘Eagle’).
Numerous gaming boards have been found varying in size: some were 8 x 8, 8 x 7, or 9 x 10, but the most common size is a grid 12 x 8 squares. We may not know which size of board was used for what game, but the size probably did not affect the style of play. Boards for all such games may have been interchangeable. Boards were made of wood, but some were made of stone, marble, or even silver.
Gameplay: Except for the special characteristics of the ‘King’, Latrunculi has these rules:
Use a 12 x 8 board with the starting arrangement as shown right.
Black plays first.
Pieces may move any number of spaces in a horizontal or vertical direction.
A single piece is captured if surrounded on two opposite sides.
The board edges cannot be used as an aid to capture pieces.
A piece in the corner can be captured by two of the opponent’s pieces placed across the corner.
Multiple pieces can be captured along a line.
The Rex (Dux or Aquila) cannot be captured but can be immobilised by being surrounded on all four sides.
The first player to immobilise the opponent’s ‘King’ wins.
The ‘King’ is immobilised if it is blocked by an opponent’s piece such that it has no place left to move.
If the game stalemates, the player with the most captured opponent’s piece wins.
Sequences of plays that repeat endlessly are to be avoided. This is usually obvious to both players after two series of repeated moves - any move initiating a third repeating series of moves is prohibited.
Players must announce when they move one of their pieces in-between two of their opponent’s pieces to avoid any later dispute.
Tabula
Tabula, meaning a ‘plank’ or ‘board’, was a Greco-Roman board game for two players and is thought to be an ancestor of modern backgammon. The earliest description of ‘τάβλι’ [1] is in an epigram of Byzantine Emperor Zeno (r. AD 474 - AD 475; AD 476 - AD 491), given by Agathias of Myrine (6th-century AD). In it Agathias describes a game in which Zeno goes from a strong position to a very weak one after an unfortunate dice roll.
The τάβλι of Emperor Zeno's time is believed to be a direct descendant of the earlier game Duodecim Scripta (also known as Ludus Duodecim Scriptorum; see right). Tabula, however, has the earlier board's middle row of points removed, with only the two outer rows retained.
Popular with soldiers, Tabula reached Arabia by Roman expansion into the Middle East in the 1st-century AD. Tabula appears to have spawned a series of games throughout Europe, such as Ad Elta Stelpur in Iceland, Taefle and Fayles in England (AD 1025), Sixe-Ace in Spain (AD 1251), and Tourne-case in France. The Arabian game Nard appears to be a slightly modified version of Tabula, perhaps incorporating aspects of Egyptian Senet. Nard spread to the Far East in about AD 220 and became widely popular.
Gameplay:
The board, as illustrated right, is similar to a modern backgammon board with 24 points, 12 on each side.
Each player has 15 pieces, all of which start off the board before entering from square 1 and moving anticlockwise round the board.
The roll of three dice scores [2] determine the moves of between one and three pieces. One piece can move the total of the three dice scores, or three pieces can each move according to the value shown on one die.
Any part of a throw which cannot be used was lost, but a player must use the whole value of the throw if possible.
If a player’s piece lands on a point occupied by an opponent’s single piece, the opponent’s piece was removed from the board. Modern interpretations require the removed piece to re-enter the game on the next throw [3].
If a player has two or more pieces on a point, then this position is closed to the opponent and these pieces cannot be taken.
No player may enter the second half of the board until all their pieces have entered the board.
No player’s pieces may exit the board (‘bear off’) until all that player’s pieces have entered the last quarter [3]. This means that if a single piece is taken, then the remaining pieces are frozen in the last quarter until the removed piece re-enters the board and catches up with them.
Bear
‘Bear’ appears to be one in a category of two-player board games popular in the Roman Empire. In this variation, the game is played with three ‘hunters’ and one ‘bear’ on a patterned board. The hunters’ goal is to hem in the bear and block its movements; the goal of the bear is to avoid this fate.
Gameplay: The initial positions for the gaming pieces may vary, but usually the ‘bear’ is placed in the middle of the board, and the three ‘hunters’ are placed together on one edge of the board.
Both ‘hunters’ and the ‘bear’ move one space at a time following the pattern on the board. Only one piece may be moved by each player. Players alternate their turns. There are no captures in this game.
As an optional rule, if the ‘hunters’ cannot stalemate the bear in a given number of moves, for example 40 moves, the ‘bear’ wins.
Tali
Tali, commonly known as ‘knucklebones’, was a game inherited from the ancient Greeks who had originally made the gaming pieces from the large anklebones (Latin: talus, pl. tali; Greek: astragalus, pl. astragaloi) of hoofed animals such as sheep, goats or cattle. While preserving their natural shape the Romans proceeded to make tali from various different materials including silver, gold, bronze, gemstones, marble, glass, terracotta, wood, ivory, and of course, bone. The tali themselves have four distinct sides each of which was inscribed with symbols, most likely Roman numerals, and each having a different value of 1, 3, 4 and 6.
Gameplay: Four tali were dropped from a moderate height over the ground or a gaming table. Scoring in the game varied widely but the simplest version only counted combinations known as ‘Venus’, ‘Senio’, ‘Vultures’ and ‘Dogs’ (see the table of possible scores right). All other results were ignored so, if a score was not achieved, the tali were thrown by the next payer and so on until a winner was decided. If money was being wagered, then the winner could claim the pot. However, if a ‘Senio’ was thrown by two or more players, a draw would be declared and the next round of throws would begin again with the pot intact.
‘Venus’, ‘Vultures’ and ‘Senio’ took precedence over all other variations of numerical score. The simple totalling of a score may have been replaced with throwing pairs as is this case in some modern card games such as poker. Other rules might be agreed upon at the start of a game in much the same way as they are at the beginning of a poker match.
The extended scoring, shown tabulated above right, is based on simple numerical superiority except for ‘Venus’, ‘Senio’, ‘Vultures’ and ‘Dogs’. In the case of values, poker-like rules prevail. In other words, a triple beats a pair, but two pair beat a triple. More advance players might adapt other poker-like rules to suit.
Surprisingly, only one ambiguity occurs with the possible scores shown above right. With a score of 14, the highest pair (‘sixes’ in this case) are assumed to beat the other highest pair, namely of ‘fours’.
And finally, in a variation of Tali played by the Emperor Augustus, anyone throwing ‘Dogs’ had to add four coins to the pot. The first player to throw ‘Venus’ would take all.
Tropa
Tropa was a variation of Tali which involved throwing the knucklebones into the narrow neck of a ceramic or glass jar. Typically three or four astragali or tali were thrown and only the score of those that entered the jar would count.
Terni lapilli
Terni lapilli, meaning ‘three pebbles’, is a two-player game of ancient Roman origin dated to approximately the 1st-century BC. Thanks to its simplicity, the game’s popularity spread widely. Game boards have been found throughout the Roman Empire scratched on walls, floors, and in public spaces. Further afield, similar game boards have been found in ancient Egypt as early as 1440 BC, in ancient Troy and Athens, and at a Bronze Age site in Ireland. Some sources claim the game was played as far afield as China from 500 BC.
Terni Lapilli is the forerunner of 'Noughts and Crosses' (US: ‘Tic-Tac-Toe’) but given that no X's and O's are scribed on the surviving game boards strongly suggests gaming pieces were used. Presumably two sets of different coloured pebbles or counters were played on either square or circular boards.
Gameplay: Beginning with an empty board the players take turns placing each of their four gaming pieces on the board. The winner is whoever first achieves three gaming pieces in a row. Sometimes the rule is used at the beginning of the game that the first player cannot place the first piece on the centre mark.
Merels
Merellus was the Latin name for 'game piece', which may have been corrupted in English to 'morris' and given rise to several similarly named games. The popular Roman game Terni Lapilli (‘Three Pebbles’) is considered the ancestor of ‘Three Men’s Morris’ which, by extension produced ‘Six Men’s Morris’, popular in Italy, France and England during the Middle Ages but obsolete by 1600, and ‘Nine Men’s Morris’ or ‘Merels’. There is even a ‘Twelve Men’s Morris’ that adds four diagonal lines to the board and gives each player twelve gaming pieces.
These games are also known as ‘Mill’ or ‘Morris’ in English and as ‘Mérelles’ in French, ‘Morels’ in Spanish, ‘Mühle’ in German, ‘Mølle’ in Norwegian, ‘Linea’ in Italy and ‘Luk Tsut Ki’ in China. As for the English names, it is believed they are derivations of the French word ‘merel’ which means marker itself derived from the Latin merellus.
The game is played on a board consisting of three concentric squares connected by lines from the middle of each of the inner square's sides to the middle of the corresponding outer square's side. Pieces are played on the corner points and on the points where lines intersect so there are 24 playable points. Accompanying the board there should be 9 black pieces and 9 white pieces usually in the form of round counters.
The basic aim is to make ‘mills’ - vertical or horizontal lines of three stones in a row. Every time this is achieved, an opponent's piece is removed, the overall objective being to reduce the number of opponent's pieces to less than three or to render the opponent unable to play. To begin the board is empty. A coin toss decides which player will play white as white moves first and thus has a slight advantage.
Gameplay is in two phases:
To begin, players each take turns to place a piece of their own colour on any unoccupied point until all eighteen pieces have been played. After that, turns alternate and consist of a player moving one piece along a line to an adjacent vacant point.
During both phases, whenever a player achieves a ‘mill’, that player immediately removes from the board one piece belonging to their opponent that does not form part of a ‘mill’.
If all the opponent’s pieces form ‘mills’, then an exception is made and the player is allowed to remove any one piece.
It is only upon the formation of a ‘mill’ that a piece is captured, but a player will often break a ‘mill’ by moving a piece out of it and then, in a subsequent turn, return the piece back thus forming a new ‘mill’ and capturing another opponent’s piece.
Captured pieces are never replayed onto the board and remain captured for the remainder of the game.
The game is finished when a player loses either by being reduced to two pieces or by being unable to move.
Sometimes a ‘wild’ rule is played for when a player is reduced to only three pieces. In this case, any player with only three pieces remaining is allowed to move from any point to any other point on the board regardless of lines or whether the destination point is adjacent.
Endnotes:
1. The word ‘τάβλι’ is still used today in Greece but it refers to backgammon. This applies as well in Syria and Turkey (as tavla), Bulgaria (as tabla) and in Romania (as table). ▲
2. Modern backgammon uses two dice and a doubling cube, the latter aimed at increasing the winning stakes. ▲
3. It is not known whether players had to re-enter 'hit' pieces before playing those on the board, nor whether players had to gather all pieces in the fourth quadrant before ‘bearing off’. It is also not clear whether there was a central ‘bar’ as in backgammon. However, if Tabula was derived from the earlier Ludus Duodecim Scriptorum, which did not have a ‘bar’, then it seems reasonable that Tabula did not have one either. ▲