Check out Alice Chess, our featured variant for June, 2024.

(The following is a revised version of the article that appeared in the March 30, 1977, issue of J'ADOUBE!, The Cincinnati Chess Magazine.)

Kriegspiel - Cincinnati style

by David Moeser

During the 1960s and the Fischer Wave era of the 1970s it seemed that every couple of years there was a local revival of Kriegspiel, egged on by such stalwarts of "fun" chess as Howard Goodrich, Mike Juhasz, Bruce Zimov, the legendary Mike Thayer, and of course, this writer. (The chessplayers of those times would undoubtedly put the emphasis on the "of course"!)

Kriegspiel is a chess variant. Simply stated, it's the playing of a game of Regular Chess without either player ever seeing the opponent's moves or pieces. The players have to cleverly guess where the opposing material is located.

Until Double Bughouse came along around 1971, Kriegspiel was the only generally accepted variation of Chess. Now that Double Bughouse has (regrettably) passed by the wayside, Kriegspiel has a clear path to counterattack its way back onto the chess scene. Its greatest advantage is that it's the only form of chess that's a true spectator sport! It's enjoyed by both participants and onlookers. It's also tremendously funny most of the time. In part that's because the players one often sees tackling Kriegspiel are novices at it. But there is skill in the game. When an Expert learns what it's all about, he plays Kriegspiel like an Expert. Among its devotees have been such Experts (and city champions!) as Bob Timmel, Mike Thayer, and Rea B. Hayes.

The name Kriegspiel, which literally means "war game," has no special significance except as a name for this particular variant of chess. Kriegspiel uses a regular board, a regular set, and the regular rules of chess. It's different only in that White doesn't formally see and know Black's moves, and Black doesn't formally see and know White's moves.

The ideal arrangement is for the players to sit back to back, at separate tables, facing away from each other, with a third table in between. On the third table a Referee places his own set, maintains the correct position at all times, and keeps the gamescore. The players must not be able to see each other's boards or the Referee's board. It's very important that spectators not move the pieces on the Referee's board or in any other way interfere in the game. It's vital to the success of Kriegspiel for the Referee's board to always be absolutely correct and for the players not to benefit from any outside information.

Over the years we've seen several different versions of rules for Kriegspiel. CHESS SUTTON COLDFIELD, the English magazine, distributed a pamphlet with their rules. Kenneth Harkness, who authored the national rulebooks for USCF in the 1950s and 1960s, explained the game in his books. But just as Cincinnati is known for its distinctive chili, so too may it be known for the following Kriegspiel rules. This Kriegspiel recipe gives the players a sufficient amount of information from the Referee to make the game a sensible one, without providing too much. These rules also avoid wasting time, as do some versions, by giving out too little information.

A few sample games conclude this article. We reprint here an epic battle, appropriately played on November 11, 1971, the 53rd anniversary of the ending of World War I, between two reigning local Kriegspiel champions. In the White trunks was the city Kriegspiel Champion, Rea B. Hayes. In the Black trunks was the University of Cincinnati Chess Club's Kriegspiel King, Mike Juhasz. This game was played at the Parkway Chess Club, which at that time met at the Central Parkway YMCA in Cincinnati, Ohio.

The rules

The idea of Kriegspiel is this: The player on move attempts either to make a legal move, or to make moves called "tries" which have as their prime objective the obtaining of insight as to what the real position is. Any move or "try" which really is a legal move stands as that player's actual move in the game.

On his own board, each player's own pieces are "official," and the Kriegspiel version of "touch move" applies to them. They're for real, and their position must be identical to that shown on the Referee's set. The opposing pieces have no formal standing. Each player may set up pieces of the opposing color anywhere he wants, or not set them up at all; and he may move them any time he wants. (The Referee must completely ignore opposing-color material on each player's board. It isn't "official" and it may be on "wrong" or even absurd squares. The Referee must not allow himself to be confused by it.)

The Kriegspiel "touch" rule applies only to each player's own material: The Player isn't allowed to touch his own material except for the purpose of making a try. That is, "trying" to make a legal move. If a "try" is legal, it stands as the Player's real move. If it isn't a legal move, the Player is not required to move (or try) that piece again.

The best procedure is for the Referee to require players to acquit (let go of) the piece they make a "try" with, and to say nothing until they do so. This requirement eliminates the kind of sticky situation where the Referee blurts out some information but the player says "Oh, I didn't really mean to play that move!" and insists on the right to take it back because he's still holding onto the piece.

After each legal move is made, and before the Opponent begins his turn, the Referee records the move, makes it on his own set, and makes the announcements required by these rules. It's important that the Referee announce all the information required by the rules -- and nothing else. All announcements must be heard by both players. Also, spectators must not talk openly about the game, for even casual comments can give away valuable information. Finally, the Referee must endeavor to be 100% correct, or else the game is likely to be ruined.


If a "try" is a legal move, the Referee simply announces that the Player has moved. "Black has moved," he says. Or "White to move." After a while the Ref's likely to abbreviate this notification to monosyllables: "White." "Black." "White," he'll say.

However, if a "try" isn't a legal move, the Referee says "No" or "Illegal," and that try must be retracted. If the move is impossible, and the player must know it for some reason (like trying to capture one's own pieces), the Referee's appropriate response is "Nonsense." The "Nonsense" announcement discourages a Player from wasting time or attempting to use the Referee to mislead the Opponent.

The player continues to try to make a move until he finds one that's legal. When a legal move is completed, the Referee announces whichever of the following information is appropriate:

  1. If a capture has been made, the fact of the capture and the square the captured piece is to be removed from. (Keep this wording in mind for an en passant capture.)
  2. Whether the captured material was a "pawn" or a "piece" -- but if a piece, not what kind of piece.
  3. If a check has been made, the fact that the Opponent is now in check, and the direction of check with respect to the Opponent's King: whether that King is in check on a file, on a rank, on the long diagonal, on the short diagonal, or by a Knight.
    1. If any of the pawns of the Player on move can capture anything, the Referee announces that the Player now has a "pawn capture," which means that at least one of his pawns can legally capture something (anything) of the Opponent's. The Referee does not tell where on the board any such capture is. The Player may now try to make captures with his pawns, or he may not. He may make tries with pieces, then tries (either possible moves or possible captures) with pawns, go back to pieces, then go back to pawn tries, etc.
    2. Due to the pawn's unique capturing power (its capturing "vector" is different from its move), a Player who attempts pawn captures when the Referee has not announced there are any is trying nonsense. In this instance the Referee announces "Nonsense" so the Opponent isn't unfairly confused. (For example, the Opponent would be led to believe the Player still has a lot of material on the board when he really doesn't.)
    3. Knotty point: If a Player is in check and has a pawn capture on the board, but no such pawn capture will remove the check, does the Referee announce the pawn capture? Answer: no.
    4. As long as at least one pawn capture is on the board, the Referee must announce that a pawn capture is possible -- and continue to do so every time the Player on move has a pawn capture available. (Note: This is because announcements apply only to the move on which they're announced.)

The following rules elaborate on certain prohibitions:

  1. If a piece (non-pawn) can capture something, that is not announced.
  2. Promotion of pawns is not announced. Each player should be supplied with extra "promotion material" at the beginning of the game.
  3. The Referee may not give a count of material on the board during the game.
  4. On the Referee's board such moves as castling or pawn promotion must be made silently and without any noticeable delay, so as not to reveal the nature of these unique movements. It's in the players' interest to do likewise. This is why the players should have promotion material available at the start of the game!

Some chess dealers sell tiny magnetic chess sets with boards no more than three inches wide. Pieces on such sets are typically just checkers with chess diagram symbols. These sets are useful for Kriegspiel referees if an extra set isn't available, or if portability is desired.


Playing the game is only half the fun. The other half is the postmortem, when the players play over the game and see what really happened! Since it ruins the party to have to stop in the middle because the gamescore can't be figured out, correct notation is important. For best results the Referee should have an assistant who records the moves and makes sure that King checks aren't overlooked. In the following game Black's play is outclassed by White's. Nimzovitch should've been a Kriegspiel player, because in Kriegspiel overprotection is essential. Note White's careful preparations and overall technique.

In descriptive notation

REA B. HAYES (X) vs. MIKE JUHASZ (B), November 11, 1971.
 1. P-KB4     P-KB3     29. R1-Q5     P-N4
 2. N-KB3     P-QB3     30. PxP       PxP
 3. P-KN3     K-B2      31. Q-KB2+    K-N3
 4. B-R3      P-K3      32. NxBP      RxN
 5. O-O       P-QN3     33. RxR       RxP
 6. P-QB3     B-N2      34. QxR       B-K3
 7. P-R3      P-N3      35. R-R5      K-B3
 8. P-QN4     P-KR3     36. Q-Q2      B-B4
 9. B-QN2     B-N2      37. Q-Q5      N-K2
10. P-Q3      P-Q3      38. P-B5      B-B1
11. QN-Q2     P-R3      39. P-B6      B-R6
12. P-B4      Q-K2      40. R-QB5     B-B1
13. P-K4      N-Q2      41. Q-B4      K-N2
14. P-Q4      R-Q1      42. P-B7      K-N3
15. R-K1      R-R2      43. RxP+      K-B3
16. P-K5      QPxP      44. R/N-Q5    K-K3
17. QPxP      PxP       45. Q-Q4      N-B3
18. NxP       NxN       46. R-Q8      B-N2
19. BxN       BxB       47. P-B8=Q+   K-K2
20. RxB       P-B4      48. Q-KB4     BxQ1
21. PxP       PxP       49. RxB       N-Q5
22. Q-K2      B-R8      50. R5-B7+    K-K3
23. R-K1      R-QB1     51. R-B6+     NxR
24. BxP+      QxB       52. RxN+      K-Q4
25. RxQ       B-N7      53. Q-B4+     K-K4
26. R-K5      B-R6      54. Q-K6+     K-Q5
27. N-K4      B-Q2      55. R-B4+     K-Q6
28. R-Q1      R-B3      56. Q-K4+     Resigns

In algebraic notation

REA B. HAYES (X) vs. MIKE JUHASZ (B), November 11, 1971.
 1. f4        f6        29. R1d5      g5
 2. Nf3       c6        30. fxg       hxg
 3. g3        Kf7       31. Qf2+      Kg6
 4. Bh3       e6        32. Nxc5      Rxc5
 5. O-O       b6        33. Rxc5      Rxh2
 6. c3        Bb7       34. Qxh2      Be6
 7. a3        g6        35. Ra5       Kf6
 8. b4        h6        36. Qd2       Bf5
 9. Bb2       Bg7       37. Qd5       Ne7
10. d3        d6        38. c5        Bc8
11. N1d2      a6        39. c6        Bh3
12. c4        Qe7       40. Rc5       Bc8
13. e4        Nd7       41. Qc4       Kg7
14. d4        Rd8       42. c7        Kg6
15. Re1       Rh7       43. Rxg5      Kf6
16. e5        dxe       44. Rgd5      Ke6
17. dxe       fxe       45. Qd4       Nc6
18. Nxe5      Nxe5      46. Rd8       Bb7
19. Bxe5      Bxe5      47. c8=Q+     Ke7
20. Rxe5      c5        48. Qf4       Bxc8
21. bxc       bxc       49. Rxc8      Nd4
22. Qe2       Ba8       50. R5c7+     Ke6
23. Re1       Rc8       51. Rc6+      Nxc6
24. Bxe6+     Qxe6      52. Rxc6+     Kd5
25. Rxe6      Bb2       53. Qc4+      Ke5
26. Re5       Ba3       54. Qe6+      Kd4
27. Ne4       Bd7       55. Rc4+      Kd3
28. Rd1       Rc6       56. Qe4+      Resigns


The author would like to add two comments on the concepts of "nonsense" and "any." First, I've noticed that some Kriegspiel enthusiasts either avoid applying the "nonsense" announcement rule, or reject it. The result is that a player can get away with non-sensical "tries" all day long and the referee acts as that player's dupe by going along with the ruse. I believe that good referees should be able to understand why this point is important, should be sharp enough to spot nonsense tries, and should rigorously apply the rule.

Secondly, one school of Kriegspiel believes in the "Any?" rule, whereby a player has to ask whether there are any possible pawn captures. Anybody who's played Kriegspiel for more than three minutes knows that under this system the proper, and necessary, thing to do is to ask "Any?" on every single move, even when it's obviously unlikely, illegal, or impossible. The result is inquiries by both players, on every move, lasting the entire game, and sometimes repeated several times during a turn to move. The opponents of the "Any?" rule consider all this to be very silly.

The worst of both worlds is when the "Any?" rule is used and the "nonsense" announcement rule is not! That turns the game into a joke. In contrast, the rules suggested in this article make for a practical and reasonable version of "blind" chess, with an encouragement for a significant element of skill -- like regular chess.

A reader's comment

A reader, called `Sam' commented that there may be situations where it is very hard for a referee to correctly decide whether to call `nonsense' or `no', and he suggests, at least in competive play, to avoid the nonsense call and always use the `no' call for impossible moves.

He also asks for computer programs that take the role of referee.Apart from some internet servers, I do not know these. Please inform me about such programs.

See also:

Written by David Moeser. Last part by Hans Bodlaender.
WWW page created: June 25, 1997. Last modified: January 23, 1999.