Etymology and origins 
The Arabic word shatranj is derived from the Sanskrit chaturanga (catuḥ="four", anga="arm"). In Middle Persian the word appears as chatrang, with the 'u' lost due to syncope and the 'a' lost to apocope, e.g., in the title of the text Mâdayân î chatrang ("Book of Chess") from the 7th century AD. In Persian folk etymology, the word is sometimes re-bracketed as sad ("hundred") + ranj ("worries"), which might appear quite meaningful to players. The word was adapted into Arabic as shatranj, and then into the Portuguese xadrez, Spanish ajedrez, and Greek ζατρίκιον; but English chess and check come via French échecs (Old French eschecs) from Persian شَاه (shāh = "king").
The game came to Persia from India in the early centuries AD. The earliest Persian reference to shataranj is found in the Middle Persian book Karnamak-i Artaxshir-i Papakan, which was written between the 3rd and 7th centuries. This ancient Persian text refers to Shah Ardashir I, who ruled from 224–241, as a master of the game:
By the help of Providence Ardeshir became more victorious and warlike than all, on the polo and the riding-ground, at Chatrang and Vine-Artakhshir, and in several other arts.
However, Karnamak contains many fables and legends, and this only establishes the popularity of chatrang at the time of its composition.
During the reign of the later Sassanid king Khosrau I (531–579), a gift from an Indian king (possibly a Maukhari Dynasty king of Kannauj) included a chess game with sixteen pieces of emerald and sixteen of ruby (green vs. red). The game came with a challenge which was successfully resolved by Khosrau's courtiers. This incident, originally referred to in the Mâdayân î chatrang (c. 620 AD), is also mentioned in Firdausi's Shahnama (c. 1010).
The rules of chaturanga seen in India today have enormous variation, but all involve four branches (angas) of the army: the horse, the elephant (bishop), the chariot (rook) and the foot-soldier (pawn), played on a 8x8 board. Shatranj adapted much of the same rules as chaturanga, and also the basic 16 piece structure. There is also a larger 10x11 board derivative; the 14th-century Tamerlane chess, or shatranj kamil (perfect chess), with a slightly different piece structure.
In some later variants the darker squares were engraved. The game spread Westwards after the Islamic conquest of Persia and a considerable body of literature on game tactics and strategy was produced from the 8th century onwards.
With the spread of Islam, chess diffused into the Maghreb and then to Andalusian Spain. During the Islamic conquest of India (c. 12th century), some forms came back to India as well, as evidenced in the N. Indian term māt (mate, derivative from Persian māt) or the Bengali borey (pawn, presumed der. Arabic baidaq). Over the following centuries, chess became popular in Europe eventually giving rise to modern chess.
|Fers or Wazīr (Counsellor or Ferz)|
|Rukh (Chariot or Rook)|
|"Pīl" in Persian and "al-Fīl" in Arabic (Elephant)|
|Asb (Horse in Persian) or Knight|
|Sarbaz (piyadeh) (Pawn)|
The initial setup in shatranj was essentially the same as in modern chess. However the position of the white shah (king), on the right or left side was not fixed. Either the arrangement as in modern chess or as shown on the diagram above were possible. In either case, however, the white and black shāh would be on the same file (but not always in modern India). The game was played with these pieces:
- Shāh (king) moves like the king in chess.
- Fers (counsellor; also spelled ferz; Arabic firz, from Persian فرزين farzīn ; also called Wazīr) moves exactly one square diagonally, which makes it a rather weak piece. It was renamed "queen" in Europe. Even today, the word for the queen piece is ферзь (ferz) in Russian, vezér in Hungarian, "vezir" in Turkish and "vazīr" in Persian. It has analogue to the guards in xiangqi.
- Rukh (chariot; from Persian رخ rokh) moves like the rook in chess.
- Pīl, Alfil, Aufin, and similar (elephant; from Persian پيل pīl; al- is the Arabic for "the") moves exactly two squares diagonally, jumping over the square between. Each Pīl could reach only one-eighth of the squares on the board, and because their circuits were disjoint, they could never capture one another. This piece might have had a different move sometimes in chaturanga, where the piece is also called "elephant". The Pīl was replaced by the bishop in modern chess. Even today, the word for the bishop piece is alfil in Spanish, alfiere in Italian, "fil" in Turkish, "fīl" in Persian and слон (which means elephant) in Russian. The elephant piece survives in xiangqi with the limitations that the elephant in xiangqi cannot jump over an intervening piece and is restricted to the owner's half of the board. In janggi, its movement was changed to become a slightly further-reaching version of the horse.
- Faras (horse, from Arabic; Persian اسپ asp) moves like the knight in chess.
- Baidaq (from Arabic بيدق from Persian پياده piyāda, foot-soldier, by adapting the Persian word as Arabic bayādiq, which was treated as a broken plural from which was extracted an apparent singular baidaq) moves and captures like the pawns in chess, but not moving two squares on the first move. When they reach the eighth rank, baidaqs are promoted, but only to fers.
Pieces are shown on the diagrams and recorded in the notation using the equivalent modern symbols, as in the table above. In modern descriptions of shatranj, the names king, rook, knight and pawn are commonly used for shah, rukh, faras, and baidaq.
There were also other differences compared to modern chess: Castling was not allowed (it was invented much later). Stalemating the opposing king resulted in a win for the player delivering stalemate. Capturing all one's opponent's pieces apart from the king (baring the king) was a win, unless your opponent could capture your last piece on his or her next move, then in most parts of the Islamic world it was a draw, but in Medina it was a win.
Early Arabic shatranj literature 
During the Golden Age of Arabic, many works on shatranj were written, recording for the first time the analysis of opening games, chess problems, the knight's tour, and many more subjects common in modern chess books. Many of these manuscripts are missing, but their content is known due to compilation work done by the later authors.
- Al-Adli's Kitab ash-shatranj ('Book of chess')
- Ar-Razi's Latif fi'sh-shatranj ('Elegance in chess')
- As-Suli's Kitab ash-shatranj (two volumes)
- Al-Lajlaj's Kitab mansubat ash-shatranj ('Book of chess-positions or problems')
- B. Aliqlidisi's Kitab majmu'fi mansubat ash-shatranj ('Collection of chess problems')
There is a passage referring to chess in a work said to be by Hasan, a philosopher from Basra who died in 728 AD. However the attribution of authorship is dubious.
Player classification 
Al-Adli as well as As-Suli introduced classifications of players by their playing strength. Both of them specify 5 classes of players:
- Aliyat (or aliya), grandees
- Mutaqaribat, proximes - players who could win 2-4 games out of 10 in the match against grandee. They received odds of a pawn from grandee (better players g-, a- or h-pawn, weaker ones d- or e-pawn).
- Third class - players who received odds of a fers from grandee.
- Fourth class - received odds of a knight.
- Fifth class - received odds of a rook.
To determine his or her class, a player would play a series or match with a player of a known class without odds. If he won 7 or more games out of 10, he belonged to a higher class.
- Jabir al-Kufi, Rabrab and Abun-Naam were three aliyat players during the rule of caliph al-Ma'mun.
- Al-Adli was the strongest player during the rule of caliph al-Wathiq. At this time he was the only player in aliyat category.
- Ar-Razi in 847 won a match against an already old al-Adli in the presence of caliph al-Mutawakkil and so become a player of aliyat category.
- As-Suli was the strongest player during the reign of caliph al-Muktafi. Ar-Razi was already dead and there were no players of comparable strength before as-Suli appeared on the scene. In the presence of al-Muktafi he easily won a match against a certain al-Mawardi and thus proved that he was the best player of that time. As-Suli considered Rabrab and ar-Razi as the greatest of his predecessors.
- Al-Lajlaj was a pupil of as-Suli and also a great shatranj master of his time.
Game play 
Openings in shatranj were usually called tabbiyyaتَبِّيّة (pl. tabbiyyaat),تَبِيّات which can be translated as battle array. Due to slow piece development in shatranj, the exact sequence of moves was relatively unimportant. Instead players aimed to reach a specific position, tabiya, mostly ignoring the play of their opponent.
The works of al-Adli and as-Suli contain collections of tabiyat. Tabiyat were usually given as position on a half-board with some comments about them. The concrete sequence of moves to reach them was not specified. In his book Al-Lajlaj analyzed some tabiya in detail. He started his analysis from some given opening, for example "Double Mujannah" or "Mujannah - Mashaikhi", and then continued up to move 40., giving numerous variations.
Piece values 
Both al-Adli and as-Suli provided estimation of piece values in their books on shatranj. They used a monetary system to specify piece values. For example, as-Suli gives piece values in dirhem, the currency in use in his time:
The chess pieces were often shaped to avoid any representation of a real living thing.
As-Suli also believed that the b-pawn was better than the f-pawn and King's side Alfil (on the c-file) was better than Queen's side one (on the f-file). Furthemore, an Alfil on the c-file was better than the d-pawn and the Alfil on the f-file was better than an e-pawn.
c. 10th century
Persian chess masters composed many shatranj problems. Such shatranj problems were called mansūba (pl. mansūbāt). This word can be translated from Arabic as arrangement, position or situation. Mansubat were typically composed in such a way that a win could be achieved as a sequence of checks. One's own king was usually threatened by immediate checkmate.
One Mansuba is the Dilaram Problem shown at the right. Black threatens immediate checkmate by 1...Ra2# or Ra8#. However, white can win with a two-rook sacrifice:
- 1.Rh8+ Kxh8 2.Bf5+ Kg8 3.Rh8+ Kxh8 4.g7+ Kg8 5.Nh6#
Note that the Alfil (bishop) moves two squares diagonally, jumping over intermediate pieces; this allows it to jump over the white knight to deliver the discovered check from the second rook with 2.Bf5+. It was said that a nobleman wagered (playing white) his wife Dilārām on a chess game, and this position arose, and she appealed "Sacrifice your two Rooks, and not me."
See also 
- Unknown court historian of the Sassanid Empire (before 628 AD). The Karnamik-I-Ardashir, or The Records of Ardashir. http://www.fordham.edu/halsall/ancient/ardashir.html. Note: Vine-Artakhsir is a reference to the game later known as Nard, a predecessor of Backgammon.
- Murray, H.J.R. (1913). A History of Chess. Benjamin Press (originally published by Oxford University Press). ISBN 0-936317-01-9.
- Jean-Louis Cazaux (12 March 2004). "The Enigma of Chess birth: The Old Texts: 6th, 7th and 8th centuries". Retrieved 14 July 2007.
- Jean-Louis Cazaux (16 June 2006). "Indian Chess Sets". Retrieved 14 July 2007.
- A History of Chess, bottom of p.311, by H.J.R.Murray, publ. Oxford at the Clarendon Press.