2000-06-08 last update, 1996-09-23 first day, Robert Jasiek

Super Ko

A board play may not recreate a position.

Introduction

After the primitive ko rule super ko is the simplest of all rule sets, that restrict moves to ensure a finite go game end. Super ko is only one rule, which prohibits repetition of whole board position. Strictly speaking, super ko is no ko rule because no ko is defined, but a restriction rule. The above rule is the standard variant of super ko, called positional super ko. Other variants of a restriction rule exist. They are said to belong to the repetition class of ko rule sets. Examples are available.

Major Variants

primitive ko: A move may not recreate a position.

positional super ko: A board play may not recreate a position.

situational super ko: A board play may not recreate a position previously left by the player.

natural situational super ko: A player may not use a board play to recreate a position if he has used one to create it.

fixed ko: A board play may not recreate a pair of positions.

Comments

The primitive ko rule is only of game theoretical interest: The first player without a legal move loses. Hereby the only available move type is board play; primitive rules do not use pass plays.

A move is either a pass play or a board play. A pass play gives the other player the right to move. A position is a whole board together with its configuration of stones. Recreation refers to any prior position during a game and can never occur, because each of the rules prohibits it. A player leaves a position by finishing the performance of one of his moves. I.e. super ko refers to positions after completion of moves.

The actual difference of positional super ko and situational super ko as to occuring positions is rather small. Very few distinguishing examples are known. With its longer text situational super ko allows slightly more variation. With positional respectively situational super ko repetition analysis refers to a position respectively a position together with the right to move.

The natural situational super ko rule treats causes and consequences as to prohibition the same way: Both can only arise due to board plays. Thus natural situational super ko is conceptionally clearer than situational super ko that also allows pass plays to cause situational prohibitions. In practice, natural situational super ko and situational super ko have almost always the same outcome, although natural situational super ko sometimes allows unfamiliar move sequences.

Fixed ko is a multiple repetition rule: the position before and the position after a board play are considered together and in order for prohibition. So not recreation of just one position is prohibited but recycling. As a practical result, no ko fight at all occurs. Any ko is a so called disturbing ko. Hence strategy and tactics, incl. life and death, are severely altered. The rule is of great theoretical importance, however, it could also be misused for simplified go.

Comparion

Positional Super Ko

+ simplest text
+ easiest to explain in comparison with basic ko plus exceptions
+ evaluates visible information only
+ concept used in NihonKiin49, WAGC79, Chinese88
+ widely known as the super ko rule
o in practice almost the same behaviour as situational super ko
o there is exactly one known rare special position with unfamiliar behaviour
- does not make an attempt to maximize variation

Situational Super Ko

+ makes some attempt of increasing variation
+ concept used in NewZealand, AGA91
o in practice almost the same behaviour as positional super ko
o non-existence of doubtful special position not shown
- longer text
- refers to a combination of visible and invisible information
- inconsistency: causes of prohibiton due to board plays or pass plays, application of prohibition only for board plays

Natural Situational Super Ko

+ makes some attempt of increasing variation
+ consistency: causes and application of prohibition both refer to board plays only
o after some pass play a few classes of positions have unfamiliar move sequences compared to positional and situational super ko
o a few classes of positions have other behaviour than positional resp. situational super ko
- longer text
- refers to a combination of visible and invisible information
- not yet used in practice

History

1900 - 1968: Rule theory was discussed especially in Japan and the USA, but I have no access to sources, in which super ko might have been described.
1945: Edward Lasker: "Modern Chess Strategy" (appendix on go): situational super ko.
1968-09 - 1969-11: Ikeda Toshio rules: positional super ko. Since the Japanese 1949 rules were illogical, Ikeda proposed in a series of eleven articels in Igo Shincho logical rule sets. 1992 an English version was published: "On the Rules of Go".
1974-10: Republic of China Wei-ch'i Association: positional super ko. 1952 Taiwanese had adopted a translation of the Japanese 1949 rules. Probably Ikeda's criticism led to a new set of rules drafted by a rules committee. The text was lengthy. 1975 an English translation appeared: "The Laws of Wei-ch'i".
1975: New Zealand Go Organization: positional super ko. This was a shortened text of the Taiwanese version. Before no English text of the Japanese 1949 rules was available and an attempt of writing them down failed due to their illogic.
1978: New Zealand Go Organization: situational super ko. The first version of New Zealand rules was developed.
1991: American Go Association: situational super ko. Adoption of a logical rule set including area rules was the occasion for changing to super ko.
1995: Tromp-Taylor rules: situational super ko. John Tromp invented a simple and logical rule set for go, that was revised by Bill Taylor and published in the newsgroup "rec.games.go". The Tromp-Taylor rules are virtually the simplest of all logical go rules, that describe the flavour of go, and are also useful for beginners.
1996-09: Tromp-Taylor rules: positional super ko. Since there is only a slight difference to situational super ko, the simpler variation has been implemented.
1998: The author invented the natural situational super ko rule.
1998: The author invented the fixed ko rule.

Alternative

A super ko rule is neither used to justify any special position nor to maintain exceptionally void games. Traditionalists may regard this as a disadvantage. However, the only logical flaw is the possibility of move-sequences of exponential length depending on the number of board points. This is generally known as the problem for players to recognize repetition with many kos on the board. The Basic Ko Rules are a possible alternative by allowing linear length in all practical cases.

Acknowledgements

To Ikeda Toshio for promoting the simplest useful ko rule. To Matti Siivola for proposing the terms "positional" and "situational". To all having provided historical information. To all promoting super ko.