The history of shogi

From chaturanga to Heian shogi

Shogi belongs to the chess family, arriving in Japan from the Asian mainland sometime before the eleventh century — probably via China and South-East Asia rather than directly. The earliest documented Japanese form, Heian shogi, is recognisable but smaller and slower than the modern game.

What happened next is unique in the chess family: instead of one game, Japan produced a whole series of them, and they got larger. Chu shogi is played on a 12×12 board with 46 pieces a side and is a genuinely great game still played today. Dai shogi is bigger again, and the tradition runs to absurd extremes — tai shogi on a 25×25 board with 177 pieces a side.

The invention of drops

The modern game won out because of a rule change that appears around the sixteenth century: captured pieces may be returned to play by their captor. No other member of the chess family has it.

The traditional explanation is that it reflects the reality of Sengoku-period Japanese warfare, where defeated soldiers routinely changed allegiance rather than being killed. Whether or not that is the actual reason, the effect on the game is decisive: it turned a 9×9 board with mostly short-range pieces — which would otherwise be a slow, drawish game — into the sharpest of all the chess variants.

The Edo period: shogi becomes a profession

Like Go, shogi was professionalised under the Tokugawa shogunate. From 1612 the government paid stipends to shogi masters, and hereditary houses — the Ohashi and Ito families — held the title of Meijin, passing it down by succession rather than competition.

The Edo period also produced shogi's great problem literature. Tsume shogi — forced mating problems where every move is check — became an art form, and two collections from the period, Shogi Zuko and Shogi Muso, contain problems of extraordinary length and elegance that are still studied. Solving tsume is still the standard training method for reading ability.

The hereditary Meijin ended in the twentieth century. The Japan Shogi Association was founded in 1924, and from 1935 the Meijin title was contested competitively — the change that created modern professional shogi.

Habu, Fujii and the modern professional game

Professional shogi in Japan is a major spectator sport. Games are sponsored by newspapers, broadcast with commentary, and the top titles — Meijin, Ryuo, Osho, Kio, Oi, Oza, Kisei and others — are contested in long matches held at traditional inns, with the players' meals reported as news.

Yoshiharu Habu dominated the 1990s and 2000s, and in 1996 achieved something no one else has: holding all seven major titles simultaneously. He remains the benchmark by which professional careers are measured.

Sota Fujii, who turned professional in 2016 at the age of fourteen, began his career with a record-breaking unbeaten streak and has since rewritten the record books for youngest title holder at almost every level. His rise coincided with — and was helped by — the arrival of strong engines as study tools.

Why shogi took longer than chess

Shogi resisted computers for two decades after chess fell, and the reason is drops. Because any piece in hand can be played on any empty square, the branching factor is far higher than in chess — often over a hundred legal moves, and more as the game goes on rather than fewer.

The turning point was the Denou-sen series of matches from 2012 onwards, pitting professionals against programs. Programs took the series in 2013, and in 2017 the engine Ponanza defeated the reigning Meijin, Amahiko Sato, in a two-game match. Engines are now clearly stronger than any human.

As in chess, the aftermath has been productive rather than destructive: professionals study with engines, opening theory has been rewritten — engines rehabilitated several "bad" formations — and the game's popularity has risen rather than fallen.

Common questions