Fairy chess is a generic term which was introduced before the First World War to cover all
problems which are not directmates. Over time, as selfmates and helpmates grew in
popularity they became classed as orthodox, so nowadays the term fairy chess covers
problems which contain one or more unorthodox features, usually a piece or condition, less
frequently an unorthodox board.
Fairy chess is very popular with composers because of the unlimited scope offered for
achieving originality. In other genres the risk of anticipation, especially for the
inexperienced composer, is high, but in the realm of the unorthodox imagination has free
rein. The danger for the composer is that he may produce a problem which is technically
brilliant but too complicated to be entertaining for the solver. The best fairy problems
feature strategy peculiar to the pieces and/or conditions employed, and do not simply
repeat effects which could be achieved in an orthodox problem.
The genre is much too large to cover comprehensively in a short article, so I will give
examples of some of the more popular fairy pieces and forms.
Pieces:
Fairy chess enthusiasts classify pieces into three groups: leapers, riders and hoppers
(but be warned – some pieces do not fit into any of these categories!). Leapers are
pieces which play a move of a specific length, ignoring any intervening piece. Taking the
side of a square as one unit, the knight is a leaper whose move is length root 5 (think
Pythagoras!). Riders is fairy terminology for line-pieces, such as queens, rooks
and bishops. Hoppers can only move by jumping over an intervening piece, called a
hurdle.
The Grasshopper moves on the same lines as a queen, but must jump over a hurdle to
land on the square immediately beyond. Grasshoppers are shown as inverted queens and the
letter G is used in notation.
(1) Brian Stephenson & Michael McDowell
Comm., StrateGems, 2001
Mate in 3 Grasshoppers c3, c8: e1 b7
The key of No.1, 1.Ga1, threatens nothing , but waits for Black to create a
weakness. In the three variations which follow, White uses his bishop to
create a hurdle for the grasshopper. After 1...b4 White continues 2.Bc3, and
Black cannot prevent 3.Gd4 mate. Had White tried to play Bc3 before the pawn
blocked b4, Black could have defended by moving a grasshopper to b4, to meet
Gd4 with Ge4. After 1...Gb1 White continues 2.Be5 and 3.Gf6 mate. This would
not have worked earlier because Black had the defence 2...Ge6, cutting the c8
grasshopper's guard of h3 and allowing the king to escape. Finally 1...Gb4 is
met by 2.Bg7 and 3.Gh8 mate, the grasshopper being unable to return to b7 from
where it could interpose at h7.
The Nightrider plays one or more knight moves along a straight line as a single move. A
nightrider at a1 on an empty board can move to b3, c5, d7, c2, e3 or g4. Nightriders are
shown as inverted knights and the letter N is used in notation.
The convention regarding promotion in problems featuring unorthodox pieces is that
promotion is possible to any piece present in the diagram.
(2) Pierre Monreal & Jean Oudot
2nd Prize, Themes-64, 1964
Mate in 2 Nightriders b7, g3, h7: b5
The key of 2 is 1.d8N, which pins the c6 pawn to threaten 2.Qxb5. Black can defeat
the threat by moving the nightrider from b5, and this leads to a number of
variations involving interference with other black pieces. 1...Na7 interferes
with the rook, allowing the nightrider from g3 to mate at a6. 1...Nd6
interferes with the queen, and the key piece mates by capturing at c6. 1...Nd4
interferes with the other rook for 2.Qc4, exploiting the fact that the
nightrider at h7 pins the d5 pawn. 1...Nf3 interferes with the bishop, the h7
nightrider delivering another pin-mate at d5. 1...Nh2 again interferes with
the queen, allowing the g3 nightrider to deliver the pretty mate 2.Nxh1.
There are a few by-play variations: 1...Ra5, self-blocking and allowing the b7
nightrider to mate at d3; 1...Nxa3 2.Bxa3, 1...Rxd8 2.Na6, and 1...Rc4 2.Qxc4.
Chinese pieces are a family of pieces developed from Chinese Chess. The Mao and Pao
have equivalents in the Chinese game, while the Leo and Vao are derived from the Pao. The
Mao moves like a knight, but is not a leaper. It moves first one square laterally
then one diagonally; if the lateral square is occupied the Mao cannot move. The Pao
moves like a rook, but to capture must jump over a hurdle. A Pao standing at a1 with a
piece (of either colour) at a5 could move along the rank or to a2, a3 or a4, and capture
any enemy piece which stopped at a6, a7 or a8. The Leo and Vao are similar
to the Pao, but move on queen and bishop lines respectively. Chinese pieces are shown
rotated a quarter-turn to the left and the abbreviations used for them are M, P, Le and V.
(3) Brian Stephenson
The Problemist Supplement, 2001
Mate in 2 Leos g6; g1: Paos e1, f1 Maos d3; e2, f3
3 illustrates the sort of strategy possible with Chinese pieces. The key. 1.Rc7,
threatens 2.Lec6. Black can defend by playing a piece to d4, cutting the Mao's
guard of c5 in anticipation of the Leo cutting the rook's guard of the same
square. If 1...Led4, White mates with 2.Lexg8, which would have been
impossible earlier, as the black Leo would have recaptured. After 1...Mfd4
2.Lef7 is mate because the Mao's move has left the f1 Pao without a hurdle and
consequently it cannot capture the Leo. 1...Med4 2.Lee6 repeats this strategy.
Finally 1...d4 opens a line for 2.Bxg8.
Boards:
Cylinder boards were very popular in the first half of the 20th century. The
vertical cylinder has the a and h-files joined, while the horizontal cylinder has
the 1st and 8th ranks joined. The combination of both is called the anchor ring or torus.
(4) Artur Mandler
Prager Presse, 1928
Mate in 3: Vertical Cylinder
In 4 the missing board edges indicate a vertical cylinder board. The problem
illustrates a typical mate in cylinder chess. The key 1.Kf2 places the king on
the only square where it will not interfere with lines that White needs to
keep open, as the variations will show. 1...Ka8 is met by 2.Bd2, which guards
b8 via h6, and forces the king to h8. White then mates by 3.Qh1 (moving via a2).
Examination of the squares in the king's field shows that the queen guards a8,
h7 and g8 (via a2), while the bishop guards g7 (via a5) and a7 (via h6). The
other variation ends in a mate in which the placing of the white pieces and
the guards are similar to those of the first mate. 1...Kc8 2.Bh2 (via a3) Kd8
3.Qd1 (via h5). Such a duplicated mate is called an echo.
Unorthodox boards often feature in less serious problems. 5 is famously known as
“The Space Queen”.
(5) Thomas R. Dawson
British Chess Magazine, 1943
Mate in 7
As it stands the problem is an orthodox mate in 7. After the key 1.Bb1, the
queen continually pins the knight while moving ever closer to the king. The
solution runs 1...Kd1 2.Qd6 Kc1 3.Qf4 Kd1 4.Qd4 Kc1 5.Qe3 Kd1 6.Qd3 Kc1 7.Qc2
mate. Dawson realised that if the board was extended upward and outwards
beyond the h-file the manoeuvre could be extended indefinitely, and calculated
that if a board with 2 inch squares was assumed and the problem reset as a
mate in 69 – “the white queen would start 240000 miles away from c1 - ON
THE MOON IN FACT – and we have a vision of a lunar queen sweeping
astronomically through space in a tremendous zig-zag path, converging
remorselessly to strike the black king to his doom - silver Fate swooping
down.” (T.R. Dawson, Caissa's Fairy Tales 1947)
Conditions:
New fairy conditions are constantly being invented by creative composers. Many fail to
catch on, some enjoy a short vogue then fade away, but others become firmly established.
One of the most successful novelties is Circe Chess. The basic premise is that a
piece which is captured is replaced on its starting square. If the square is occupied then
the piece is removed as in normal chess. There are a few rules governing replacement: a
rook, bishop or knight returns to the starting square of the same colour as the square on
which it is captured; a pawn returns to the starting square of the file on which it is
captured. The king is exempt in ordinary Circe; if the king is also subject to the Circe
condition the problem will be labelled “Circe Rex Inclusiv’.
(6) Norman A. Macleod
1st HM., Europe-Echecs, 1977
Mate in 2: Circe
In 6 discovering check from the bishop is futile, as the black bishop
will interpose at d5, and after 2.Bxd5 it will reappear at c8 and interpose at
b7! The key is 1.Kb4, which threatens nothing but clears the a-file, so that
any capture on e1 will send the rook to a1, pinning the bishop. After
1...dxe1Q+ (Ra1) White can recapture safely 2.Sxe1 (Qd8). 1...dxe1R cannot be
met by 2.Sxe1, as the rook will be reborn at h8, allowing 2...Rxh1 (Bf1).
However, White can mate with 2.Sg1, shutting out the rook. After 1...dxe1B+
(Ra1) 2.Sxe1 is illegal, as the bishop will be reborn at f8, giving check, but
as the bishop cannot interpose on the long diagonal 2.Sd2 is mate. 1...dxe1S
(Ra1) attacks the long diagonal, a situation which would not be altered by
2.Sxe1 (Sb8), so this time 2.Qe4 mates. The remaining variation 1...B anywhere
2.Ra7 illustrates a peculiarity of Circe, namely that a piece can guard
itself! 2...Kxa7 is illegal as the rook will be reborn at a1.
In Madrasi Chess a piece when observed by a similar piece of the opposite colour is
paralysed, and can neither move, capture nor check, but may paralyse in turn. As with
Circe kings are exempt in the normal form, and any problem in which the king is also
subject to the condition will be labelled ‘Madrasi Rex Inclusiv’.
(7) Norman A. Macleod
Springaren, 1984
Mate in 2: Madrasi
Some typical Madrasi effects are illustrated in 7, where the rooks at c5 and
c8 are initially paralysed. The key move 1.Sb4 does not deliver check, as both
knights are paralysed, but threatens 2.Qxd5, unparalysing the white knight.
Note that 2.Bxd5 is not a threat because the bishop becomes paralysed,
allowing the king to escape to b3; similarly 2.Rxd5 would allow a flight at
d2. Black can defend by attacking d5 with his queen, so that after 2.Qxd5 the
king can capture on d1, which will no longer be guarded. 1...Qe6 interferes
with the black bishop, allowing 2.Bxd5, while 1...Qd6 interferes with the
rook, allowing 2.Rxd5. 1...Qc6 unparalyses the c4 rook, leading to 2.Rcxd5.
Seriesmovers are problems in which one side makes a unique sequence of uninterrupted moves
after which the other makes a single move, fulfilling the stipulation. The most popular
type of seriesmover is the serieshelpmate, in which Black makes n moves to reach a
position where White can mate in 1. Black must not move into check, and may only deliver
check on the last move of the sequence.
(8) Josif Krikheli
1st Prize, feenschach, 1966
Serieshelpmate in 25
8 illustrates a variety of strategies commonly found in serieshelpmates,
namely king-walk, line-closure and promotion. The arrangement of pieces above
the white king is currently stymied. As the g3 pawn will never move, the rook
cannot emerge from h4. The c-pawn can promote, but only to a bishop or knight.
The black king cannot be mated in the corner to which it is confined by the
queen, so it must be extracted and a mate organised on the queen's-side.
Looking around for a possible mating move we eventually consider Qa6, mating
the king at a5 with b4 blocked by a bishop gained by promotion. To release the
king, the h3 bishop must reach f7, but that requires the promoted bishop to
take over the task of shielding the white king from check. After the king
emerges the bishops swap roles, with the black-squared bishop helping the king
to reach a5. The full solution runs 1.c5 2.c4 3.c3 4.c2 5.c1B 6.Be3 7.Bg1
8.Bh2 9.Bf1 10.Bc4 11.Bf7 12.Kf8 13.Ke8 14.Bc4 15.Bf1 16.Bh3 17.Bg1 18.Bc5
19.Be7 20.Kd8 21.Kc7 22.Bd6 23.Kb6 24.Ka5 25.Bb4 Qa6 mate.
Construction tasks are not strictly speaking problems, but positions which demonstrate
some record feature. Many construction tasks involve one move, for example demonstrating
the maximum number of possible moves, checks, or mates. Many such tasks come in two
versions, featuring legal or illegal positions.
