Friendship: Gerd Wilhelm Hörning & Gerhard Josten

by Siegfried Hornecker

The friendship between Gerd Wilhelm Hörning & Gerhard Josten was forged by endgame studies and their column that offered insights in the process of correcting endgame studies. Bonding a love for our art over a study by Leonid Kubbel, they soon composed on their own.

As the Nobel Prize in Physics in 2022 told us, everyone might live in an own reality, and as such I can only account for what I believe happened in my reality. The violation of Bell inequalities can only be resulting in the complete disproval of locality, thus concluding there is an own reality for everyone.

Around 2004 when I was still a new endgame studies composer, Michael Roxlau from the German magazine Die Schwalbe printed me out a lot of addresses for international competitions from the website of Jan Golha, through which I got to know a few composers abroad. I don't remember however how the contact to Gerd Wilhelm Hörning and Gerhard Josten was established - likely it was about their column "Ausgerechnete: Endspiele" in the Rochade Europa, back then one of the two big German chess monthlys. Their column analyzed endgame studies, found cooks, and offered improvements whenever possible. Those were the pre-internet days for me, so we talked on the telephone and later also via e-mail, composing a version of the Saavedra study together. I remember that I contributed the two knight sacrifices, he contributed the rest but couldn't get the move 1.h6-h7 to work as a captureless key, so I found a setting for a good first move. Of course, my memory might be wrong, but that is how I remember it.

Siegfried Hornecker & Gerhard Josten, The Problemist May 2005, 5th commendation.

White to move and win

The two composers from Köln (Cologne) decided to end their column after 50 episodes, and for the final one they also invited me to present an endgame study. I chose what I believed to be a fun expansion of a known Allumwandlung (all-promotion) setting, in fact the first such by Sigurd Clausen in 1927. My addition was to add the phoenix theme, i.e. all the four pieces that could be promoted into are captured prior to the promotion of the pawn. If we see a pawn on the 8th rank as Schroedinger's piece, does this make it Schroedinger's phoenix? If so, what about his cat? Oh... wait... the cat already has died from old age even if it had enough sustenance. Similar, I might add, to our wrong belief that Harold Lommer's 1933 endgame study was the first Allumwandlung endgame study.

Siegfried Hornecker, Rochade Europa November 2004 (after Clausen).

White to move and win.

Hörning & Josten's book "Schach zwischen Krieg und Kunst" contains their slightly modified long article "Studieren geht über Probieren", published in many parts originally, and "Ausgerechnete: Endspiele", which also appeared in the same magazine Rochade Europa since February 2000. It is my main source for this article.

In their chess club, an endgame study was presented that was nearly impossible to solve. When they finally found the solution to that endgame study, the love for this genre was kindled in Gerd Wilhelm Hörning and Gerhard Josten.

According to how I understand the book, the study presented on page 25 is the one that they tried to solve but couldn't find a solution for until finally it dawned on them after days of trying...

Leonid Kubbel, 64, 30 July 1925. 1st prize.

White to move and win

The both composers from Cologne developed a fine taste for chess curiosities, including the Babson task, an incredible idea where a pawn promotion is answered by the same kind of pawn promotion, for all four kinds of pawn promotion (queen, rook, bishop, knight). Gerhard Josten based one of his works on the idea, but there was no promotion aspect. The expert solvers didn't find his intended theme, do you?

Gerhard Josten, Schach-Report October 1992.

Mate in 2

Their series "Ausgerechnete: Endspiele" was born from analyzing endgame studies invarious sources, later including often the database of Harold van der Heijden. Its second version in 2000 was distributed by ChessBase, but for later versions Harold took an approach of distributing it via paid digital download on an own website.

The column had its humorous moments but contributed to the popularization of endgame studies in Germany. Episode 6 featured the world-famous Mitrofanov Deflection. But should it be named after Mitrofanov? Hörning and Josten quoted a study by Paul Farago.

Paul Farago, Revista de Romana de Sah 20 February 1936.

White to move and win.

1.g7 Qh5+ 2.Qg5!! indeed is the same deflection as used by Mitrofanov whose correction is also quoted by Hörning and Josten.

Leopold Mitrofanov, Rustaveli MT 1967, 1st prize. USSR overall championship 1967-1968, 1st place. Correction by Mitrofanov, Vecherni Leningrad 1971

In this complicated study, the deflection appears after 1.b6+ Ka8! 2.Re1! N:e1 3.g7 h1Q 4.g8Q+ Bb8 5.a7! Nc6+! 6.d:c6 Q:h5+ and the draw is secured after 7.Ka6 Qe2+, but not after 7.Qg5!! Q:g5+ 8.Ka6! B:a7 9.c7! An incredible position! Black has queen, bishop and knight against two pawns. He is not in check. He has the right to move. And yet, the knight is his downfall, as 9.-Qa5+ 10.K:a5 B:b6+ 11.K:b6 is not stalemate. Of course there is a cool ending that would work without the knight if 9.-Qa5+ 10.K:a5 Kb7 was forced: 11.b:a7 K:a7 12.c8R (with the knight, 12.c8Q wins). So the only thing that would make this study better would be a logical try where White captures the knight. But that would be like asking someone who climbed the Mount Everest to do it again but on one foot...

Unexpected errors in sources might also lead to curiosities, as the authors would find out in their seventh issue of the series:

Harold Lommer, Manchester Guardian 30 June 1949.

White to move and draw

Believing the solution to be 1.Rc3+ Kb1 2.Ke3+ B:f5 3.Ke2 Nd1 4.Rf3 Bg4 5.Kf1 B:f3 stalemate, Hörning & Josten found the cook 1.-Kd2. In a reply, John Roycroft speculated that the solution was taken from "1357 Endspielstudien" where Lommer apparently only remembered the try but not the actual solution: 1.Ke3! B:f5 2.Rc3+ Kb1 3.Ke2 Nd1 4.Rf3 Bg4 5.Kf1 B:f3 stalemate, i.e. with the first two moves played in reverse order. So does this save the endgame study? Well, Yours Truly found that first of all the intermediate check 4.Rb3+ Ka2 is possible, but then the intended solution is the only one: 5.Rf3 Bg4 6.Kf1 B:f3 stalemate. Similarly 4.Ra3 also draws, but leads to the correct path after 4.-Kb2 5.Rf3 Bg4 6.Kf1 B:f3 stalemate. Loss-of-time duals? Or destructive?

Soon the issues of the column "Ausgerechnete: Endspiele" took up specific thematic connections between all the presented endgame studies. Issue 8, as an example, featured works by Josef Hasek.

Readers were invited to give their improvements to cooked studies and also to send their feedback to the columns. This happened quite a few times, but oftentimes studies remained uncorrected. The authors also learned a lot during their time with the column, such as threefold repetition not applying in endgame studies, after the following study was presented in their 12th issue. This earliest success by Andrey Visokosov was presented and higly praised on the "Chess Café" internet page in 2000, where our heroes saw it.

Andrey Visokosov. Shakhmatnoye obozrenye 1997, 2nd prize.

White to move and draw.

The threefold repetition at move 17 (see replayable entries) is not relevant in the setting of an endgame study. Neither are other game-only rules, such as the 50/75 move rule, in evaluatng if an endgame is won. John Roycroft reacted to the same issue with a letter in which he commentated on all of its four endgame studies.

One of the great successes of the column was that many endgame studies were corrected, such as the following one in issue 18 of the series.

Alexey Troitzky. 500 Endspielstudien, 1924.

White to move and win (cooked)

The intended solution is 1.c6 d:c6 2.d6 c:d6 3.h4 g:h3 4.g:h3 Kc5 5.h4 Kd5 6.h5 Ke6 7.h6 Kf6 8.f5 is the main variation but the cook 1.-d6! 2.h4 g:h3 3.g:h3 Ka4! shatters White's dreams. Similarly, 4.-Ka4! also works in the main variation. Setting the king from f1 to d1 is the idea of Hörning and Josten (that apparently was known before?). Mikhail Zinar set the king to a1 instead (replayable below). Both work and preserve the intended solution up to the point above. Both however fail to preserve the second part of the solution: 8.-b5 9.Ke2 b4 10.Kd3 d5 11.Kc2 d4 12.Kc1! c5 13.Kc2 c4 14.Kc1! b3/c3/d3 15.Kb2/Kc2/Kd2 wins

It seems that the following work by Hasek that was featured in issue 19 of the column had also a wrong solution given in the source used by our both heroes.

Josef Hasek, Sachmaty 1967 (likely: Shakhmaty v SSSR 1967).

White to move and win

Hasek's solution is 1.Rb8+ Kh7 2.Rg7+ N:g7 3.Rh8+ K:h8 4.f:g7+ Kh7 5.B:c2 N:c2 6.g8Q+ Kh6 7.Qg7+ Kh5 8.Qh8+ Kg5 9.Qf6+ Kh5 10.Kg7 winning. However, the source used by Hörning and Josten gave 2.B:c2 as the second move of the solution, where Black just wins with 2.-Bh5+ 3.Ke7 N:c2. Their variations given in the column are in the database of Harold van der Heijden, likely they were added after appearing in the column. So good work could even come from wrong sources.

Many more examples of curiosities could be quoted from the column, of course, but I think this short overview will be enough for the readers to grasp that while the column was not highly professional, it was created with a lot of love by the authors, trying to not only improve endgame studies but also to bring them closer to a general public - in a monthly magazine where every reader could send in articles and many were published, even a low-quality short story by Yours Truly from back in the day for which he is wholly embarassed now.

After the magazine was bought in 2014, it became more professional, but lost its touch with the chess composition world in the opinion of many people. This is regrettable, as such a small gem as "Ausgerechnete: Endspiele" now hardly would be published anywhere. For a young Siegfried Hornecker it was a monthly source of joy and curiosity in the maybe one and a half years that he read it.

The authors pursued also other avenues of chess. Gerd Wilhelm Hörning was a known name - likely still is - regarding computer chess, at least in the magazine "Computer-Schach und Spiele" which seems to be a pure online magazine nowadays. Gerhard Josten's creative spirit was expressed in beautiful paintings. Their love for composing endgame studies seems to have died down, and unfortunately Gerhard Josten stopped composing altogether after bad experiences with some people. [https://www.gerhard-josten.de/kunst/ His website] is a source of aesthetical pleasure in his artistic regard.

As a chess historian, Josten also contributes to preserve and unearth knowledge in his numerous books. Yours Truly was honored to have participated in the project "A Study Apiece" where chess composers presented one of their endgame studies, later he learnt that "Caissas Schlossbewohner" by Peter Kniest used a similar formula, but with Josten's book a generation after that the contemporary mostly German chess composers were given their voice, and since then another generation has passed...

As Gerd Wilhelm Hörning and Gerhard Josten have mostly retreated into their personal life, Yours Truly does not feel at liberty anymore to disclose any private information known to him on the both subjects of this month. Instead, here is a short overview of their endgame studies.

Including corrections, versions, etc. there are 153 studies alone or as co-author by Gerd Wilhelm Hörning and 274 by Gerhard Josten in the current database by Harold van der Heijden. One of Josten's earliest works already won the first prize in the magazine "Schach". His active years as chess composer were from 1999 to 2012. Hörning's active years as chess composer were from 1999 to 2014. The dates given here are publication dates of their endgame studies.

Hörning's approach to endgame studies was unusual, as he tended to create compositions that were made to fool either solvers or computers. With this in mind, modern hardware and software is not as easily fooled anymore, and the potential of AI and neural networks has increased tremendously since the early 2000s.

In a special match of the game show "Jeopardy!" (hosted by the always beloved Alex Trebek) IBM's AI Watson outsmarted superstar Ken Jennings easily. James Holzhauer did not have his revolutionary new approach to the show in 2013, and even if he had, likely Watson also would not have had suffered from the same psychological impact as human contestants. The IBM computer won easily, despite some flaws. This was ten years ago. Ten years in which AI evolved further and further. The text-creation AI Chat-GPT recently was unveiled and while not always being factually correct and in some cases failing to understand an instruction, usually it is very competent (see example after the end of this article). AI now can create compelling images, text, and likely a lot more. Remember that this is only what is known to the public - the conspiracy theoretician in Yours Truly says that all that technology likely is five years behind what secret services or the military can access.

Gerhard Josten's approach to composing endgame studies is more traditional. He created endgames of practical or artistical value. For both composers, I chose four of their works each to be replayed below.

At some point my contact to the both composers stopped. But when I was young, composing at first, they gave me the opportunity to participate in their endeavours on a small scale but one that greatly honored me as a chess composer. Having grown up with a lot of emotional baggage, it gave me not only a great confidence boost to be recognized and to a degree befriended (including a personal meeting in 2006) by two of my heroes, but also showed me that the world of chess composition is a very welcoming one to new composers. In a sense, the story of Hörning and Josten is mirrored in my own. As a young child, maybe six years old, I just learnt chess, but in the book "Schach für Anfänger" by Laszló Orban, after being told that king and rook against king and rook is a draw, a small endgame study was published, which - closing our circle - was in fact a correction of the author's original intention.

Georges Emile Barbier & Fernando Saavedra. Glasgow Weekly Citizen, 4 May 1895 (version: Bohemia, 1902)

White to move and win

1.c7 Rd6+ 2.Kb5 Rd5+ 3.Kb4 Rd4+ 4.Kb3/Kc3 Rd3+/Rd1 5.Kc2 Rd4!! was explained in the book in great detail, and being a quick learner I also easily grasped the given line of 6.c8Q? Rc4+! 7.Q:c4 stalemate, which originally was Barbier's intention. Spanish reverend Fernando Saavedra, who read the newspaper, made a strange discovery, which, reprinted on Orban's book, started my lifelong love for endgame studies, and chess composition in general. White has the unbelievable move 6.c8R!!

King and rook versus king and rook is a draw. But each rule has an exception. It is Black to move. The material is absolutely equal. Four pieces on the board. Yet, the threat of checkmate necessitates 6.-Ra4 and after 7.Kb3! the checkmate threat is repeated on a1, meanwhile the rook on a4 is also attacked. White wins.

I don't compose much anymore myself. Should I never have another endgame study in a FIDE Album again, I am glad that essentially the Saavedra idea - sans promotion - is one of my final ones.

Siegfried Hornecker & Martin Minski. Rezvoz-95 MT 2017. 4th prize.

White to move and win.

After 1.Bb5+ c4! 2.B:c4+ Ke1 White wins with Saavedra's double attack: 3.Bf1! K:f1 4.g4! R:g4+ 5.Kf3 (Ra4 6.Rd1 mate) - a short study that doesn't show anything new or remarkable but still made it into the FIDE Album for some reason.

The theme of our article was friendship. The friendship between Hörning and Josten, deepened by endgame studies, in the end helped Yours Truly to thrive himself as a composer. As such, this article is not only a small testament to two normal chess players just having fun with compositions and becoming contemporary artists - for Josten in multiple ways - themselves, but it also is a documentation that those two men helped a young man to gain confidence in himself, and to become a good composer as well as the author of this monthly column. A monthly column with the same intent as theirs - bringing endgame studies to the masses, enjoying them, and offering various aspects of them.

Gerd and Gerhard, many thanks for what you did for me! Thank you for having been there when I needed someone like you to exchange ideas and conversations!

Chat-GPT output example

Many chess composers started using the AI "Chat-GPT" for recreational and humoristic purposes. Andrew Buchanan talked to Yours Truly about some examples of impressive output he received. Hiroaki Maeshima, who helped us with our previous article about Japan, asked Chat-GPT to output an ode to chess composers. With his kind permission the poem is reproduced in full below. He posted it on a Discord server on 15 January 2023, which is most likely directly after it was created.

I asked ChatGPT to compose an ode to chess composers. ChatGPT composed:

===

Ode to the chess composers

Makers of puzzles that are true contenders

Their minds are ever working

Creating new positions, and always smirking

As they craft their chess masterpieces, with no pretenders.

 

With each puzzle they compose

They challenge players to expose

Their chess skills and acumen

Through their creative men

With every move, they expose.

 

Their art is one of the mind

Creating chess positions that are truly refined

Their work is a delight

For chess lovers day and night

A true treasure, one of a kind.

 

So let us raise our glasses high

To the chess composers, let us not pass by

For they bring a new dimension

To our chess obsession

And make the game, truly alive.

Links:

• All study of the month columns
• Chess Magazine Rochade Europa website