The earliest game for which there is firm evidence of a relationship with backgammon is “Ludus Duodecim Scriptorum” (“game of twelve line”), the games board contains many similarities to that of backgammon, including the a bar, with six points on either side, however it also had three rows and was played with three dice. The earliest mention of the game is in Ovid’s Ars Amorous (written between 1 BC-8 AD).
By 54 AD a game called “Tabula” which is a variant of Ludus Duodecim Scriptorum had appeared on the scene, and it is thought that Emperor Claudius may have discussed it in a book on gaming which he wrote. However none of the 40+ books which he is said to have written have survived. The only known mention of this particular book is by Suetonius so the existence of a finished book on the topic may be in some doubt.
A board from about 125 AD of Ludus Duodecim Scriptorum has been found at Holt in Denbighshire (Wales) in a grave of a roman soldier, indicating the first wave of this game spreading across the world.
The second wave of the game spreading occured from the third century onwards and appears not to be due to the movement of soldiers but rather due to settlers and traders.
The chinese text Hun Tsun Sii (written between 960 and 1279 AD) claims that the game known as t’shu-p’u was invented in western India and imported into China during the Wei Dynasty (220-265AD). The chinese variant played the game with 16 pieces. It is also thought that the game appeared in Japan under the name Sunoroko prior to Empress Jito’s death.
However as the chinese text was written possibly over a thousand years after the period it describes, it is almost certainly relying on secondary sources (given the state of archeology in China at the time it is unlikely primary evidence would have been available). Unfortunatly these sources are currently unknown.
The claim that the game originated in India seems unlikely given the apparent similarity between Tabula and the older roman game “Ludus Duodecim Scriptorum”. No archaelogical evidence has turned up to indicate that the game may have been created in India.
About the same time as the game is thought to have entered China, it appears to have enter Great Britain, and for this we have the greater evidence of having a Tabula board from the end of the third century in a cementry in Leuna, Saxony in the grave of a germanic settler. Another board, this time from the sixth century, also thought to be of germanic origins was found in Taplow, Buckinghamshire.
Given that the game was being spread around the world at this time, it is a suprise to find an unknown board game in the tombs of Qustul from c. 350 AD which matches the board used for “Ludus Duodecim Scriptorum” indicating perhaps that while Tabula hadn’t completely wiped out “Ludus Duodecim Scriptorum” as suggest by Bell.
Despite the game having been spread far and wide, no written description of the game is known until the sixth century, where we find a description of a game position that was played by Emperor Zeno (476-481AD) given by Agathias of Myrine (527-567AD) in the form of an epigram, and from this the rules of the game have been reconstructed by RC Bell. It is though that at this the neither the doublet rule or the last quartile rule existed at this time.
By the end of the sixth century the name Tabula had been replaced by Alea, a word meaning die (as in dice). This term earliest recorded use is in Isidore of Seville’s Origines where he also gives the definiton of the term, “Alea, id est ludus tabulae…” (“Alea, that is the game of tabula”). This term seem to spread rapidly with a seventh century English text using the term “alea teblae” to refer to the game.
By the start of the ninth century Tabula turned up in the Arab world under the name, Nard, perhaps agreeing with the earlier point of Qustal that Tabula hadn’t become popular in the middle-east at that same time as it grew popular elsewhere. It has been suggested that the word “nerdshir” in the Babylonian “Gemara” is the old form of the word Nard.
In the 13th century, sometime between 1251 and 1282 AD, the Alfonso X manuscript documented the game and several variants. This document is the first known codification of the rules.
Also in the late 13th and early 14th century Tables as the game was then known, was again growing popular in England, with the game being mentioned in popular literature again for the first time in over 500 years. Complete game boards from this period have been found at St Denis near Paris (c1200 AD), Gloucester (c1220 AD) and Freiburg in Germany (c 1300 AD) .
After this point the game faded out of popularity until the sixteenth century when a variant of the game similar to the game played today took hold and rapidly spread through Europe. Known originally as tric-trac the exact rules are unknown, although they may have been document at the time no such document has been found. The name was later used to refer to the game known as backgammon in England. However it appears that the Tables variant known as Irish may have preceeded this time, and may have derived from one of the early spikes of popularity of the game.
The earliest known documentation of this new variant was in Hoyle’s “A Short Treatise on the Game of Backgammon” in 1743 but it is known from 1605 that this variant included the doublet rule and from a 1668 source that it included possibility of wining a double game (a gammon).
The first known use of the word Backgammon was in 1645, however at this time the name “Tables” was still in use and it wasn’t until the end of the seventeenth century that the name Backgammon came into standard use.
By the 18th century Backgammon had arrived in America, with Thomas Jefferson listing gambling losses for the month of June 1776 in his notebook.
Some histories of backgammon claim that Tabula, and by descendancy, Backgamon, is based on the game Senet. There are two problem with this theory:
1) No Senet boards have been found from between 1500BC-200AD. It’s possible that boards in this period will be found, but as things stand there is a signicant gap between Senet and Tabula (a ‘missing-link’), if one game came from the other we would expect both to be played at the same time with one variant slowly dominating.
2) Some arguments use the similarity of rules as a basis for a relationship for the games. However it is important to remember that we have no record of any rules of Senet, all of our information is based on limited references in period literature and paintings depicting the game. The rules most commonly cited are those of RC Bell, which were created _assuming_ the game was related to Tabula, and so arguements along this line fail due to circular reasoning.
It is possible that Tabula is based upon “The Game of Thirty Squares”, the same arguments against it apply as Senet, but Thirty Squares does a have a similar board to “Ludus Duodecim Scriptorum”. However it lacks the bar and the similarity of board may just be coincidental.
Footnotes: Suetonius in his “Lives of Caesar” talks about Emperor Claudius(died 54 AD) playing it.  RC Bell states that the book was entirely about Tabula, and cites his source as Suetonius. However according to the 1883 translation of Suetonius work by Alexander Thomson the actual statement was,
“He was fond of gaming, and published a book upon the subject. He even used to play as he rode in his chariot, having the tables so fitted, that the game was not disturbed by the motion of the carriage.”
If we assume by “tables” it is meant the game Tabula then it likely is mentioned in the book, however note the book as whole is not specifically said to be about that one game. Bell states it’s in the National Museum of Wales.  As far as I can tell no English translation of this document exists, so I’m using Bell as a secondary source for this info.  Now I assume in the location of modern day Pakistan.  Bell also accounts some information about the game in Japan, however he doesn’t indicate if this information came from Hun Tsun Sii or from another source.  Does anyone know if Hun Tsun Sii listed any sources ?  Of course this could be due to the fact that while a large amount of historical research has been done in Rome and to a lesser extent the Middle-East/Near-East, the politically unstable condition of the northern India/Pakistan area has meant only a limited amount of research has been possible.  Board is now at the British Museum. Exhibition code M&ME 1883,12-14,7.  Board is now at the British Museum. Exhibition code M&ME 1883,12-14,7.  Sourced from Emery’s Nubian Treasure  Bell refers to Myrine in Asia ????  I’ve been unable to find the original, so again I’m sourcing from Bell.  Bell’s seventh rule, opens with the line “an additional rule not mentioned by classical writers…”, does this indicate that Bell was aware of some other relavent period writings that he doesn’t tell us about ?  After throwing a doublet (i.e both dice the same) you can move as if you got that dices value four times.  All pieces have to be in the last quarter of the board before any piece can bear off.  Sourced from Bell, however I’m not certain if this it true, certainly a number of the later names for the game (tablas, tables,etc.) appear to be derived from tabula, however these could be independant inventions coming from the premise that the game was played on a table.  Sourced from Oxford English Dictionary, Second Edition.  Bell states the original is in the monestary of St. Lorenzo del Escorial.  My understanding is that substantial parts of the manuscript were based on previous arabic works, however no arabic works on the game have been found from this period to the best of my knowledge.  Robert Mannyng’s “Brunne’s Handlyng Synne” in 1303, the OED list cites from 1297, 1300, 1330 and 1386, including one by Chaucer.  Bell says seventeenth century, but it appears he wasn’t aware of the English author Forrest who in his “Grysilde” referred to the game “pastyme at Tables, Tick-tacke or Gleeke” in 1558. See also .  This name is used in the Middle East, Germany and France, and was also used in England for a while, following the principle of an spreading invention keeping it’s name, the name and by deduction the game appeared to have originated in France.  I’ve seen a reference to 16th century book, Panurg’s “Le Tric-trac clericorum”, but I haven’t been able to track this book down.  Several histories of backgammon including that of Bell seem to think Backgammon and tric-trac the same game. But there is clear evidence that they were seperate games, for instance in Sedley’s Bellamira (1687) there is the line “I lost three sets at back-gammon, and a tout at trick-track”.  There are several reference to Irish existing before this new variant, see notes  and .  As far as I can trace the only first edition remaining is at the Harry Ransom Humanities Research Center.  David Jones’ “Wittes pilgrimage” (1605) included the verse,
The Tables of those Hands LOVES Tables are;
“Hir fingers are the Points, both whitest Ware:
Whose Sides are edged with the sweetest Aire,
So to distinguish them, more sweete, pure, faire!
Tick-tack plaies, or at Queens-game;
But, Irishe hates, for hauing Trickes too blame:
Here Hee casts Doublets, Double Points to take,
(The Hart, and Hand) both which an end do make
Of all LOVES Games, saue when the Vies are paid” Thomas Shadwell’s “The sullen lovers” (1668) Act III,
“I was playing at Back-Gammon for my Dinner, which I won; and from thence we came to five up for half a piece; of the first set I had three for love and lost it: of the second I Gammon’d him, and threw Doublets at last, which you know made four, and lost that too; of the third I won never a Game.”
Note that this is the earliest known use of the word gammon. In 1645 James Howell wrote in his “Letters” (Volume 2, I believe, but I can’t confirm) that “Though you have learnt to play at Baggammon, you must not forget Irish, which is a more serious and solid game.”  For instance in Samuel Pepys’ Diary (1665).  Used in the anonymous “The Dutch Rogue (1683)”, James Howard’s “The English mounsieur” (1674), Sir George Etherege’s “She wou’d if she cou’d” (1671), Thomas Shadwell’s “The miser” (1672)  Including, an Egyptian satirical papyrus (c. 1100 BC), a wall drawing at the tomb of Baqt at Benihasan (c. 2000BC), another in the tomb of Rashepses at Sakkarah(c. 2500BC) and a wall painting in a tavern in Pompeii (c. 70AD). Note that in none of these are we sure of the identity of the game being played.  The first arguement is weaker here, for we have a board of a variant, “Game of Twenty Squares” etched onto a Colossal winged bull from sometime in the 8th century BC and another one from 12th century BC made of ivory.  For instance if one applied the same reasoning elsewhere, we would have that Alquerque is related to Seega.  The 2nd Ed OED’s entry for Gammon is incorrect in listing the earliest use as 1735.  Now in the British Museum, exhibit code EA 10016  Now in the British Museum, exhibit code GR 1897.4-1.996  Sourced from Oswald Jacoby & John Crawford’s “The Backgammon Book”  Source: Ian Riddler’s “When there is no end to a good game” in British Archeology, Issue no 31, February 1998.