Introducing: Reformed Courier-Spiel
by Clément Begnis

This is the result of my personal attempt to reform the Courier-Spiel proposed by H.G. Albers in 1821 (which was itself a modernized version of the medieval Courier Chess, my favorite historic chess variant).


Note: the chess fonts used in the diagrams of this page and in the ZRF implementation were designed by Jean-Louis Cazaux for his book L'Odyssée des jeux d'échecs (see references below). He kindly allowed me to utilise them.

Why create this game?

Originally, I wanted to make the medieval Courier game more dynamic by introducing modern moves and conventions. Up until now (and as far as I know) there have been only two historical attempts to modernize this game: Albers's Courier-Spiel and the Modern Courier Chess invented in 1992 by FIDE Master Paul Byway (in fact there have been others very little known variants but they were not openly presented as modernized versions of the medieval game).

I was not statified with Paul Byway's game for several reasons. Firstly, he calls bishop the medieval courier that moves like our modern bishop and gives the name courier to the alfil (the medieval bishop). This is of course a minor issue but I think it is important to keep the name courier for the orthodox bishop to respect the game's history, since this courier is hypothetized to be the first proto-bishop in European chess history (the medieval players found this piece so surprising and powerful that they named the game after it). Secondly, a nagging issue is how Paul Byway enhances the alfil's power (it can now leap in eight directions). While this piece should indeed be stronger in a modernized variant, this new move does not correct the problem there is in the opening position of the medieval game: the pawns that stand on b- and k-columns are unguarded. Byway does not define castling. Instead, an unmoved king is given the power of taking a double step to a vacant square. This is in my view unfortunate because it makes the game look more medieval than modern (it calls to mind the different extended leaps of kings in the European medieval assizes). But what bothers me the most is that he adds two fers instead of the medievals counsellor (or sage or man) and Schleich (or spy or jester or fool). Even if an unmoved fers can leap like a courier into a vacant square in order to speed up the game, these pieces remain very weak and slow compared to the others (and even weaker than the medieval pieces they replace!). As a result, the army which now has a modern queen in its ranks looks unbalanced and, most important, the dynamic of play, a cornerstone of the modern game, is hampered. In this respect, I need to mention here that at first (in the 1970's), Byway had made an attempt to modernize the Courier game with more powerful extra pieces. He wrote that “it was an error” because “the searchlight glare of powerful pieces wash[ed] out any subtle shades” (cited from Variant Chess Magazine, see exact references below). I will not discuss this point, but I really think that a game which claims to be “modern” needs powerful pieces. Of course, all this is just my point of view. Unlike Byway, I don't have a great knowledge regarding the theory of the game, I'm just a chess history enthusiast. That's why, even though I'm not satisfied with his proposal, I have much respect for his work.

The other attempt that had been made to modernize Courier Chess, the Courier-Spiel (by H.G. Albers from Lünebourg), is far older since it goes back to 1821. It's my favorite because Albers succeeded in introducing some modern moves while we can clearly see that the game comes from the medieval Courier Chess. Firstly, the medieval Schleich is given the moves of the medieval counsellor to make it stronger (that is to say, it steps one square in any direction) and to differentiate this new Schleich that he calls queen's counsellor or fool from the medieval counsellor, he enhances the power of the later which now combines the king's moves with the knight's. This is a brilliant idea: the pieces are stronger but we can see where the new moves come from. Then, another good point is the way he improves the move of the alfil that he calls bishop: in addition to its medieval leap, it can now make one single step diagonally so as to protect the pawns I was talking about earlier. By doing so, he successfully protects all the pawns in the initial setup without changing the position of the pieces in comparison to the original game. It's all the more interesting that, moving this way, the alfil becomes as strong as a knight. Finally, like Byway, Albers gives their modern moves to queens and pawns (initial double step + en passant capture). Castling is defined in this way: “When castling, the king moves to the square on the c-column or j-column; the rook goes to the nearby square jumping over the king (so to d1, i1, d8, or i8.) In addition to the restrictions on castling of normal chess, the king may also not castle when the rook moves over an attacked square or when the rook is attacked. So, for instance, when white wants to castle with his rook on a1, then neither king nor the rook on a1 may have moved, and none of the squares a1, b1, c1, d1, e1, and f1 may be under attack by an enemy piece when castling” (cited from, see exact references below); and the definition of pawn's promotion is even more curious: “After a pawn reaches the last row, it must stay there immobile for two moves, only at the third turn after the movement of the pawn it may move as the piece it is promoted to” (ibid). I think this odd rule comes from the Freudensprünge rule of the medieval game (see Murray's A History of Chess for more informations, most notoriously p.392).

So what do I propose?

Albers's Courier-Spiel is in my opinion an almost perfect way to modernize Courier Chess, so I'm going to base my proposition on this version. It just needs some minor, but necessary, changes to make it fully compatible with our modern conventions. So here are the things that I propose to change:

=>Regarding the name of the pieces. I call archer the piece that Albers calls bishop (and that Byway calls courier). The reason is that it is the closest translation for the German word Schütze (the original name of the piece in the medieval game). Like Albers, I keep the name courier for the piece that moves like a modern bishop, so as to justify the name of the game and with respect regarding the history (see above). I call Albers's queen's counsellor/fool champion so as to conjure up its superior strength in hand-to-hand combat, and his king's counsellor is given the name of paladin to evoke his superior knight status. The others pieces have the same names as in orthodox chess.

=>Regarding the starting position. Like in Byway's Modern Courier Chess, the white king is placed on the right to make the setup look more familiar to modern chess players. It follows that the respectives positions of Albers's king's and queen's counsellors need to be inverted. Kings face each others. So, when playing with the white pieces, from left to right we have: Rook, Knight, Archer, Courier, Champion, Queen, King, Paladin, Courier, Archer, Knight, Rook. The initial joy-leaps of the rook's and queen's pawns and of the queens that were supposed to be compulsory in the medieval Courier Chess are of course abolished.

=>Regarding the moves of the pieces. They are exactly the sames as in normal Courier-Spiel, except for the champion (Albers's queen's counsellor/fool). Rooks, knights, kings, queens and pawns move like in modern chess. The courier has the move of the modern bishop. The archer has the same moves as Albers's bishop: it moves one or two squares diagonally and when moving two squares it can jump over an occupied square. The paladin has the move of Albers's king's counsellor: it combines moves of a knight and king (without being hindered by check). The champion is more powerful than Albers's fool: it still moves like a king (and it is still not subject to check), but it can now jump two squares orthogonally over anything in between (dabbâba's leap). You may wonder why I enhanced its strength. A piece that moves just one step was interesting in the medieval game which has a slow tempo and a majority of weak pieces, but keeping in mind the enlarged board, my wishes to modernize the game, and the powerful new pieces added, it was better to improve its strength to increase the game's harmony and dynamics. Albers's fool was as strong as a courier. With its new leaping ability, the champion becomes definitely stronger than the courier while it is still weaker than the paladin (that was important because it stands beside the queen). Its strength is in fact quite close to that of the rook which make it a major piece. In the medieval Courier Chess, the counsellor often stayed at close quarters to protect the king that couldn't castle. Considering the modern castling that I introduce (see below) such a piece has to be given a more offensive function, hence its new leaping ability. Moreover, even beginners will easily remember this move, and it leaves room for the champion to be developed before the endgame (Byway himself wrote that the problem with the medieval counsellor was precisely that it was not mobile enough. It thus stayed hidden until the endgame). Finally, I chose an orthogonal leap because there were already too many diagonal movements with the archers, the couriers and the queens. Through repeated trial-and-error, I managed to find a suitable compromise between strength and range of action that respects Albers's game.

E.g. on the opening position, Zillions-of-Games gives these pieces values, normalized to 10 for the rook: pawn (2,3); knight (5); archer (5); courier (6,2); champion (8,8); paladin (11,2); queen (15,8). But after having played several games, I would propose these following values that in my view better reflect what is really going on: pawn (2,1); knight (5,4); archer (5,5); courier (6,6); champion (9,6); rook (10); paladin (12); queen (16,6).

=>Regarding castling. I don't keep Albers's castling. In Reformed Courier-Spiel castling is defined so as to recover the same positions as in standard modern chess: the king moves to the square on the c-column or k-column and the rook jumps over it (to the d- or j-columns). So when castling, the king always moves four squares towards the rook (whereas it moves only three squares in Albers's game when castling with the kingside rook). Most important, unlike Albers's rule there is no addition to the restrictions on castling of normal chess: the king can castle even if the rook is attacked or moves over an attacked square. All the others requirements (the king has not previously moved, etc.) are exactly the sames as in modern chess.

=>Regarding pawn's promotion. It is also different from Albers's rule since I apply here the orthodox conventions: when reaching the 8th row, the pawn is immediately promoted to any kind of piece, just as in normal chess (including here the archer, the paladin and the champion), and it doesn't have to wait for two moves like in Albers's Courier-Spiel. It was really necessary to abolish this strange rule, not only because I think it is a clumsy survival of the Freudensprünge rule, but especially because as John Gollon pointed out in 1972, a lot of things were unknown or unclear about this rule (see Eric Greenwood's page in the references below).

Finaly, concerning all the others rules that I don't mention here (stalemate, threefold repetition etc.) we will refer to FIDE orthodox chess.

Conclusion: When all is said and done, I've changed five aspects of Albers's Courier-Spiel: the names of three pieces, the initial position of four pieces, the moves of one piece and the castling and promotion rules. The goal is reached: Reformed Courier-Spiel completely agrees with all the modern chess conventions while being respectful towards its roots and the original medieval game.

A Reformed Courier-Spiel set with modified Staunton pieces (handmade board).

Close-up of the champion and paladin pieces.

Try it yourself!

For those that have the software Zillions-of-Games, I created a Reformed Courier-Spiel ZRF implementation. Feel free to download it: Download!
After extracting the archive, place the .zrf file into the Rules folder of your Zillions directory, and place the entire Reformed_Courier-Spiel folder into your Images directory.
Launch Zillions, open the Reformed_Courier-Spiel.zrf rules file and you can play Reformed Courier-Spiel using Zillions-of-Games.

NEW: Thanks to Jean-Louis Cazaux, Michel Gutierrez and Jérôme Choain, you can now also play Reformed Courier-Spiel on-line with Jocly. Enjoy!


CAZAUX, Jean-Louis, L'Odyssée des jeux d'échecs, Paris, PRAXEO Editions, 2010.
MURRAY, Harold.J.R, A History of chess, (1913), New York, Oxford University Press, 1969.

Online sources:
GREENWOOD, Eric, “Courier-Spiel”, The Chess Variants Pages, July 1997. Accessed on December 22nd, 2011.
BYWAY, Paul, “Modern Courier Chess”, The Variant Chess Website, 1992. Accessed on December 22nd, 2011.
BYWAY, Paul, “The Courier Game”, Variant Chess Magazine, Vol.5, Issue 37, Spring 2001. Accessed on January 20th, 2012.

See also:

Courier Chess and Courier-Spiel (in German)
Courier Chess by Rick Knowlton
Courier Chess by Jean-Louis Cazaux
Courier Chess by Paul Byway
Courier Chess on
Courier Chess on Wikipedia

And some totally different but interesting attempts made by amateurs to reform the old Courier game:
Courier Chess Moderno by Jose Carrillo
Courier 'de la Dama' and Furious Courier by Nuno Cruz
5 modified Courier presets by Eric Greenwood
Courier de los Combinados by Charles Gilman


If you have any questions, comments or suggestions, feel free to contact me at: begnis.clement[at] (replace [at] by @).

Page created on: December 29th, 2011.
Last modified on: July 23rd, 2014 (Jocly implementation).

Thanks to J.M. for proofreading this article.

©Copyright 2011 by Clément Begnis. All rights reserved.