Chaturanga and Shatranj

The History of Chess

Course 13 · Chapter 2 · Chaturanga and Shatranj

The first chapter gave the broad route of the game. This chapter looks at its first two stages: chaturanga in India and shatranj in Persia and the Islamic world. The evidence is thin and often late, which makes it a good lesson in how historians weigh sources. Names, dates and details are flagged "check"; the broad outline agrees with general surveys, but individual claims have not been checked against primary texts.

Chaturanga in India

The name

Chaturanga is a Sanskrit word usually translated "four limbs" or "four divisions", and in Indian sources it also names the four parts of an army: elephants, chariots, cavalry and infantry (check).

The date

Surveys place the earliest chess-like game in northern India by about the sixth century CE, with some sources suggesting the fifth and others the seventh (check). The date is a range, not a fixed point.

The pieces

The pieces reflected the army: a king, a counsellor, elephants, horses, chariots and foot-soldiers. The elephant, horse and chariot are the ancestors of the bishop, knight and rook (check each).

Not a modern game Early chaturanga was not played exactly as chess is now. The counsellor, ancestor of the queen, was a weak piece, and the elephant's move was short. Some scholars also think the earliest Indian form may have been played by four players, though this is debated (check).

From India to Persia

The game reached Persia, probably in the sixth or seventh century, where it was called chatrang in Middle Persian (Pahlavi). Several Persian texts mention it. The best known are:

Text (check each)	What it says	How to weigh it
The Karnamak of Ardashir	Describes the founder of the Sasanian dynasty as skilled at chatrang.	Often called the earliest literary mention, but the text is full of legend and its date is disputed. It shows the game was known when the text was composed, not when Ardashir lived.
The Book of Chess (Mādayān ī chatrang)	A story of an Indian challenge to the Persian court, solved by a clever courtier.	A Pahlavi text, thought to date from about the seventh century (check). Valuable, but a story rather than a record.
The Shahnameh	Ferdowsi's epic of about 1010 retells the story and adds a legend of two half-brothers who contest a throne in India.	Much later than the events, and told as legend.

A pattern to watch The stories of how chess reached Persia are consistent in one respect: they say it came from India. That agreement is useful, but note that every telling comes from a later Persian or Arabic writer, with no Indian text of the same date to confirm it.

Shatranj in the Islamic World

After the Arab conquest of Persia in the seventh century, the game passed into the Islamic world, where the Persian name was adapted as shatranj (check). It was widely played in the courts and in cities, and it produced a literature of its own: problems, study positions and treatises written from about the ninth century (check).

Shatranj piece (check each)	Move	Modern descendant
Shah (king)	One square in any direction	King
Firzan or ferz (counsellor)	One square diagonally	Queen
Fil (elephant)	Two squares diagonally, jumping over a piece	Bishop
Faras (horse)	As the modern knight	Knight
Rukh (chariot)	As the modern rook	Rook
Baidaq (foot-soldier)	One square forward; captures diagonally; promotes to the counsellor	Pawn

The shatranj queen and bishop were short-range pieces, so the game was slower than the modern one. One result was a different idea of winning: a player could also win by capturing every piece but the king, and rules about stalemate differed (check). Here is a simple way to represent the shatranj pieces and their one-step movement in code, a tool for exploring the old rules.

# Shatranj: the short-range pieces, as (row, column) steps
KING  = [(dr, dc) for dr in (-1, 0, 1)
         for dc in (-1, 0, 1) if (dr, dc) != (0, 0)]
FERZ  = [(-1, -1), (-1, 1), (1, -1), (1, 1)]    # one step diagonally
FIL   = [(-2, -2), (-2, 2), (2, -2), (2, 2)]    # two steps diagonally, leaps

def moves(pos, steps):
    r, c = pos
    return [(r + dr, c + dc) for dr, dc in steps
            if 0 <= r + dr < 8 and 0 <= c + dc < 8]

print(moves((0, 2), FIL))   # the elephant in its starting corner

The Wheat Legend

A famous story says a ruler offered the inventor of the game any reward, and the inventor asked for one grain of wheat on the first square, two on the second, four on the third, and so on, doubling on each of the 64 squares. The total is 2⁶⁴ − 1, about 1.8 × 10¹⁹ grains, far more than any kingdom could supply (check). The arithmetic is sound. The story itself is a legend about exponential growth, attached to chess in medieval sources, and no one can show that it records a real event (check).

Evidence, Weighed

Claim (check each)	Evidence	Verdict
Chess began in India	Consistent later Persian and Arabic traditions; the name and piece types point to an Indian army; early finds of pieces	The usual view, though the details are thin and the date is a range
Chess is thousands of years old	Some very early boards and pieces have been called chess, but they are not clearly chess	Unproven; treat with caution
Ardashir played chess	A late and legendary text	Not reliable for the third century
Shatranj was widely played by the ninth century	Treatises, problems and later literary references	Well supported

A British comparison The old shatranj pieces reached Britain only after they had changed. A good place to see what they became is the National Museum of Scotland in Edinburgh, which holds some of the Lewis chessmen (Chapter 3): there the counsellor has become a queen seated with her hand to her cheek, and the elephant has become a bishop (check). The word "rook" itself comes from the Persian and Arabic rukh, a link from the Islamic game to the English one (check).

Not Yet Verified The web search consulted for this chapter agreed that the Karnamak is usually cited as the earliest literary mention of chatrang while remaining legendary and hard to date, that the Book of Chess is a separate seventh-century Pahlavi text, that the Shahnameh (c. 1010) retells the story, and that the game's name passes from Sanskrit through Persian to Arabic. The shatranj piece moves and dates, the four-player theory and the wheat legend's sources come from general knowledge and have not been checked against primary sources; treat every "(check)" accordingly.

Exercises

Exercise 1

Explain the route of the name from chaturanga to chatrang to shatranj, and say what the pieces of the Indian game represented.

→ Solution

Exercise 2

Calculate the number of grains on the 64th square of the wheat legend, and on all 64 squares together. Say why this makes the story a legend rather than a record.

→ Solution

Exercise 3

The Karnamak says Ardashir played chess. Give two reasons a historian would be cautious about using it as evidence for the third century.

→ Solution

Quick Reference

Names: chaturanga (India), chatrang (Persia), shatranj (Arabic)
Old pieces: shah, ferz, fil, faras, rukh, baidaq
Texts: the Karnamak, the Book of Chess, the Shahnameh
Method: note how late each source is, and whether it is story or record