Study of the Month: To be human is…?

The rise of artificial intelligence as well as chess engines, starting with Alan Turing’s Turochamp, projects like Wilhelm, endgame tablebases, all bring up fundamental questions not only about what it means to be a composer of an endgame study or chess problem, but on a fundamental level what it means to be human. Could machines replace humans? In the 1990s The Outer Limits became known not only for their science-fiction anthology episodes but also for the narrator delivering a short, often deeply philosophical, thought at the beginning and end of each story. As an example, the 1960s episode Feasibility Study was remade and included this ending narration: “For centuries philosophers and theologians have debated what it means to be human. Perhaps the answer has eluded us because it is so simple. To be human is to choose.”

Can humans indeed choose their own future, their own destiny? Are all choices predetermined? The question of free will is a philosophical one that remains unresolved. It is beautiful to believe that humans have a free will, that the outcome of each choice is not predetermined.

The same obviously can’t be said about chess problems and studies. Here the author has a specific solution, or multiple such, in mind. In addition, there might be unintended solutions. However, all of those are predetermined by the stipulation and position. Any position on a chessboard is either won, drawn, or lost in a forced way. The practical side of chess profits from the perfect information not being available during the game, neither in positions where tablebases exist nor where they don’t exist yet. A computer with perfect access to all information would not discriminate between moves that lead to an easy draw and to a for human difficult defense, provided both have at least one way to draw. Endgame studies, by definition, should only have one way to draw or win.

Noam Elkies, Internet 1991.

White to move and draw.

Noam Elkies wants us to imagine this position happening in a practical game. 1.f6 Qb3 allows both 2.f7 Qd1+ 3.Kh6 and 2.Kh6 Qd1 3.f7 but afterwards the play is unique. And yet, in one of those both timelines White is lost while in the other one it is a draw. How is that possible?

Speaking of the Turochamp inventor, one of the central figures in codebreaking in Bletchley Park, he also invented the Turing Test, which is supposed to tell the difference between a human and a computer. Should computers be able to pass the test, would those have to be classified as human? This seems doubtful, as some humans fail at the test as well, and they aren’t classified as robots. However, many forms of media, such as Mary Shelley’s Frankenstein, see human qualities in the soul rather than in the physical body. This is mirrored also in a scene from the Star Trek movies when Captain James Tiberius Kirk names Mr. Spock as “the most human” soul he ever met. There certainly must be a quality to being human that transcends the physical body, and in spiritual texts it is called the soul. Such is the premise of that scene, of the deeper understanding that comes with it.

Nearly every religion has an origin story for humanity, and in fact worldwide not only religions but other myths as well have teachings about a great flood that came upon the Earth at one time. Not only this global story that was passed down for many generations but also our own age of technology poses many events that could end humanity. Another Carrington Event could destroy over a century worth of progress. One dangerous experiment that goes wrong, one conflict that reaches global scale, one accident even might cause the end of most life, including humans, on this planet. Nanomachines wreaking havoc, called “Grey Goo”? Improbable but not impossible. Furthermore, since many decades the idea is propagated that we might live in a computer simulation, that we all are just players in a big video game.

Would a reminder of our own mortality be enough to appreciate every day of life, at least if it is free of pain, of badness, if we find joy in small things? Would this get more to the essence of what it means to be a human? If so, we don’t need to look even far into the past to see the fragility of peace, to see how broken a world we live in.

Yet, it is my dream, my hope, that humanity can set its differences aside one day. Just like on the chessboard, so also in life we should always treat each other with respect and as equals. This is, of course, shadowed by allegations and dramas that arise – from Toiletgate to other high-level cheating accusations, we might always see top players that dislike each other.

In the small world of chess composition, not all composers like each other, but there are few who actively dislike each other. Cheating is brought to a different level here, where computer assistance not only is allowed but welcome, but in solo composing tourneys the assistance of other people is not allowed. I know personally no case where such a rule ever was provably broken. More easy to prove, especially thanks to access to computer databases with many tens of thousands of endgame studies, hundreds of thousands of chess problems, is when a work is anticipated. If many works of a composer turn out to be anticipated, accusations of plagiarism aren’t far. However, in some cases accidental recompositions were proven, and even the best composers weren’t able to avoid those in the past.

Vladimir Korolkov, Lelo 1951. 1st prize.

White to move and win

1.f7 Ra6+ 2.Ba3! Rxa3+ 3.Kb2 Ra2+! 4.Kc1 Ra1+ 5.Kd2 leads to a position very similar to one from a Selman study (the only difference is that here no black pawn is on g4, leading to slightly different play where Kd2 walks straight to g5). Korolkov wasn’t aware of the other study and had the same idea independently.

John Selman jr., Tijdschrift v. d. KNSB 1949, 1st prize.

White to move and win

1.Nf5 Re1+ 2.Kd2! Rxa1 3.f7 Ra2+ 4.Ke1! Ra1+ 5.Kf2 g3+ 6.Ke3! Ra3+ 7.Kf4 Ra4+ 8.Kg5 Rg4+ 9.Kh6 Rg8 10.Ne7! Be6 11.fxg8Q/R+ Bxg8 12.Ng6 mate

This is different from using another endgame study as a base for an own composition where the inspiration is either obvious or mentioned by the composer.

Karen Sumbatyan, Oleg Pervakov & Vladislav Tarasyuk. Internet, 14 March 2019.

White to move and win

1.Be4 f5! 2.Bxf5 d3 3.Bxd3 Rxd3 4.b6! axb6 and now a position from a 1956 study by Grzegorz Grzeban is reached, as readers will recognize what happens. 5.c7 Rd4+ 6.Kb5 Rd5+ 7.Kxb6 Rd6+ 8.Kb5 Rd5+ 9.Kb4 Rd4+ 10.Kb3 Rd3+ 11.Kc2 Rd4 12.c8R Ra4 13.Kb3 wins. Of course the 1895 study by Barbier & Saavedra was the inspiration. It is funny that Black gets rid of f5 to set up the stalemate trap but loses because he has a pawn on b6 (otherwise 5.-Kb2 would draw).

“Great minds think alike”, it is said. This hints at a deeper connection between all humans, possibly between all life, in which universally the essence of being – the soul – might be originating from a singular place or event. Monotheistic religious texts would call this origin a god. Indeed also near-death experiences often report that all souls are one, that the separation of souls is merely an illusion, as is time and even the entire material world. Philip K. Dick already told us in 1974, at a speech in Metz, that the entire universe in his opinion is just what we’d describe today as a computer simulation. However, as we all tangentially or more are able to influence what happens in this simulation, it seems to Yours Truly that we should all strive to improve the conditions of living for everyone, that we should try to be the best people we can be. Chess compositions, no doubt, bring great joy in composing, solving, or just replaying. The modern internet shows us with many Youtube channels, articles, solving competitions – such as those organized each year by ChessBase as well as ChessBase India, to name just the native ones here – and other mass events that a great interest even among practical player exists to transcend the confines of the battle of wits between two opponents and instead enjoy the beauty of the artificial composition. Endgame studies not only often could arise from games but indeed in some cases already did, and also the other way around this holds true, when composers are inspired by practical play.

Vladimir Kramnik – Viswanathan Anand, Bonn 2008, World Chess Championship game 5.

Black to move.

At the end of a combination, it seems that Black is lost. 33.-Rc1+ 34.Bf1 Nxh2 35.Kxh2 Rxf1 sees White with two dangerous queenside passed pawns. However, instead of resigning, Anand played… 33.-Rc1+ 34.Bf1 Ne3!! 35.fxe3 fxe3 and Kramnik resigned.

Yochanan Afek, Schach 2008, 2nd prize (after Kramnik – Anand, Bonn 2008).

White to move and win.

The fifth game of the World Chess Championship 2008 had an incredible ending, as seen above, which inspired Afek to distill its essence into an endgame study that offers a rich battle. Whereas after the equivalent to 1.Nd6! cxd6 2.cxd6 the game ended, the study just begins. This is left for readers to solve, or replay at the end of the article.

On the other hand, there are monumental constructions that clearly show artistic ideas unachievable in practical play, as their positions would require rather cooperative than competitive play.

David Gurgenidze. Lenin MT 1970, 4th prize.

White to move and win.

With 1.h8Q? Rg6+ 2.f6 Rxf6+ 3.Kd5 Rf5+ 4.e5 Rxe5+ 5.Kc4 Re4+ 6.d4 Rxd4+ 7.Kb3 Rd3+ 8.c3 Rxc3+ 9.Ka2 b3+ 10.Kxa3 b2+ 11.Ka2 b1Q+ 12.Kxb1 Rc1+ 13.Kxc1 Bxh8 a draw is reached. This shows why 1.Ra8+! wins: After the staircase maneuver that shows four “Novotny” sacrifices by White, i.e. sacrifices that interfere the lines of the bishop and rook, Black is unable to clear the diagonal with tempo when White hides his king: 1.-Kb7! 2.Rb8+! Kxb8 3.h8Q Rg6+ leads to 10.c3 Rxc3+ 11.Ka4!, winning.

Chess composers, just like also practical players, might stem from all kinds of backgrounds. There are rather mysterious figures (there are rumors about one Russian composer that he is rich, owning private helicopters, but nothing ever was confirmed to your author), there are people who love chess composition but rather collect than compose (Tim Krabbé, as an example, is a world-renown expert for curiosities, but has composed to my knowledge only a few works). There are generous botanical experts (Alain C. White), Schutzstaffel members whose friendships with jews are notable (Ado Kramer & Erich Zepler), nephews of war heroes (Wichard von Alvensleben, whose uncle of the same name rescued high-ranking political figures and their families from certain death), victims of war (Kubbel brothers Evgeny & Leonid, Troitzky) and purge (Platov, Arvid Kubbel), and even one incredibly talented puzzle inventor (Sam Loyd). The entire human condition is collected in the world of chess composition.

Sam Loyd, Checkmate Novelty Tourney 1903, 1st prize.

Mate in 3

How can checkmate even be reached here in three moves? Obviously Black has two flights for his king, and White has all pieces participating in the attack already (except the king which is safe on f1). This puzzle is nearly impossible to solve, but rightfully is regarded as one of the most beautiful compositions in history. As such I don’t want to take the opportunity away for potential solvers to experience it. Those who give up can see the solution at the end, but be warned that this solution is one you can’t forget so you have only this one chance at solving.

Obviously, the human condition also includes the ability to appreciate art. Many years ago, I read that a computer is unable to solve any chess problem in that sense. To elaborate on this: Certainly, it might be able to show you the lines that lead to the checkmate, if properly programmed it also can tell you the themes of the problem, but it will not be able to understand the context, the achievement of the composer, the significance of the variations of the chess problem or the aesthetical consideration of the endgame study.

Mark S. Liburkin, 64 1940.

White to move and win.

Life at its core can be seen as a challenge, a struggle, a battle. The same holds true for a game of chess, but in life our opponent might be the circumstances we were born in, the search for meaning, or even our own body or psyche that prevents us from achieving greater things. Spiritual knowledge would teach that life is worth living as an experience, with the goal of becoming a beacon of unconditional love. Religious texts would teach that life is just one important step to go to some higher plane – a heaven or paradise. Materialism believes that life is strictly physical, there is nothing after death. There is however one fundamental truth in all of those beliefs: Time will not rewind. But is that really true?

The insinuation that time travel is possible is based on anecdotal evidence, in which however usually the people who claim to have traveled through time were unable to recognize such at the moment and only realized it later. One well documented case is that of Johann Wolfgang von Goethe who once met his older self (or his younger self, when he was older) during a journey. This is sometimes construed to be a doppelganger phenomenon, but by how the story unfolds, it more seems like a time travel occurrence. This story, set near Sesenheim in 1771/1779 and emotionally connected to Goethe’s girlfriend, is told by the great poet himself in his autobiography “Aus meinem Leben. Dichtung und Wahrheit”. However, Goethe himself mentioned that this book series is in parts fictional, so it is unclear if this encounter really happened or was just thought up for that manuscript.

The Mandela Effect is another interesting phenomenon that purports the possibility that past events are changed. Many of its supposed phenomena can be explained by faulty memories or mishearings, or in the case of Forrest Gump’s chocolates talk also with a pre-release trailer. Was Mandela’s death in prison reported by the people who put him there, hoping that his movement would collapse, and then widely picked up by newspapers, while him being still alive remained only a footnote, leading to the nomer of the effect? Or did he really die in prison, but was alive later as if that never happened? Most people in Eurasia and America won’t have followed in detail what happened when in South Africa, but what do people remember who actually lived there? Was there any timeline you remember where Mandela was not your first President in the 1990s?

However, some mysteries can’t be as easily explained, especially the Thinker statue.

William Langstaff. Chess Amateur 1922.

Mate in 2

I saw this problem first when it was discussed by Tim Krabbé in his book about chess curiosities. What was the last move of Black? If he moved king or rook last, then 1.Ke6 checkmates in two moves. But if he moved the g-pawn last, that pawn can only have come from g7, so 1.hxg6 e.p. is the correct solution. Tim Krabbé remarked that depending on what White moves, Black can claim either. So the en passant would be illegal, but after 1.Ke6 Black can also castle kingside 1.-0-0.

The Codex for Chess Composition has adapted various rules for retrograde analysis (see its article 16), one of which (“partial retrograde analysis”) resolved this issue by saying that both are partial solutions, depending on what Black played last. The uncertainty is resolved prior to and not after the key, but for each possible last move a solution must exist. As such, there is one solution, but it consists of two parts.

The uncertainty principle should not apply to chess puzzles. But the problem above shows that missing knowledge of the past can have a profound influence about the present. In a world where supposedly all the information we need is just a few clicks and keys away, psychological studies have been made to show that the reliance on search engines leads to a decline in the ability to memorize things. In addition, information is not always complete or widely available. Without resorting to illegal activities or physical media, most movies are only accessible in the deep web (not to be confused with the darknet, the deep web simply is the part of the internet that is not available to the public). Usually you’ll have to sign up to a subscription service with a monthly fee to watch them, or “buy” them from a digital storefront (which means that they can be taken away due to licensing reasons later). Many videoclips that relate to the Mandela Effect however are posted on public video sites under the fair use right. As such, you can easily verify that Darth Vader says “No” instead of “Luke” when revealing his connection to the hero. As a meme nearly half a century ago, it would have made sense to say “Luke” instead when quoting the sentence in a conversation, as this adds the necessary context to identify it being a Star Wars scene (and while we’re at confusing names, of course Star Wars also was one of the names for the Strategic Defense Initiative).

Fears about AI, the Mandela Effect, life itself show one thing: Nothing is certain. The future is not, the past also may not be. But in a world where nothing is certain, it is just even more important that we really are ourselves, that we try to be the best people we are, and that we try to enjoy life to the fullest.

For over two hundred million years, the dinosaurs reigned supreme on this planet. Then on what now is the Yucatan peninsula in Mexico, a piece of rock hit the Earth. An eon ended in the blink of an eye (provided that dinosaurs blinked). We humans long for security, stability, the knowledge that even the darkest night will be perforated by the morning sun. Let me end this paragraph with another quote from The Outer Limits, one that might be fitting: Life is, at best, an uncertain path. Perhaps, we should be grateful with what we have, for even these riches may disappear along the way.

Martin van Essen, Alexander Wohl & Yochanan Afek. Avni JT 2005, 1st prize.

White to move and win.

In an e-mail, Yochanan Afek confirmed that the co-author is the Australian International Master (for practical play) Alexander Wohl. The study has what in expert terms is called “flow”, this means it plays out smoothly. Here indeed the riches disappear along the way – can you figure out how, or do you have to replay the solution?

Are we humans unique in our ability to create art? Are the other civilizations in the universe that might have similar forms of puzzles like we have in chess? For the European, the American, the Asian, the African, and maybe even for those who aren’t from our planet, the beauty of chess is an universal language. It is one thing that connects us, one thing that can bring us joy, awe, and ultimately make us want to share our experience with others, showing us that we all enjoy something unexpected but beautiful. Taking the words of another German poet, Friedrich von Schiller: “The human only fully Is human when he plays.”

But doesn’t that apply also to a chess engine then? Well, it might calculate a move, play it, but it won’t feel an emotional reaction to it. And maybe that is really what it means to be human, it might not be something we can explain in words, but something we deeply feel.

To be human is to feel chess.

PS: Reading a draft of this article, Andrew Buchanan, a retro composer and one of the three editors of The Hopper, a free problem chess magazine (when you read this article, their fifth issue with some joke problems should be published), mentioned the work of Sir Roger Penrose (the English Nobel Prize winning physicist) who showed positions such as the following one on his talks about AI.

White to move.

Humans easily evaluate this as a draw.

Yours Truly uses Deep Fritz 11, which would also draw this position, evaluating it at -28.13 pawn units but never capturing on a4 or b5 as that would lower the evaluation to a forced checkmate. The point made by Penrose is in that evaluation, as a human immediately would evaluate it as 0.00 pawn units. Of more interest is another Penrose position:

White to move.

Humans see 1.Bb4 immediately, engines want to play 1.Bxa5. It would be interesting to test this with modern engines with neural networks and Monte Carlo algorithms. It does not take long to evaluate 1.Bxa5 b4 as lost, but how long does it take to evaluate 1.Bb4 as a draw?

Roger Penrose’s love for chess might be rooted in his family: Jonathan Penrose, the brother of Roger, was a British chess grandmaster. His father Lionel Penrose, among numerous other achievements, was a chess theorist… and study composer.

Lionel Penrose, 1234 (Dover edition) 1938.

White to move and win.

The solution is short but sweet so should pose no issues to the readers.

(Note: Many of the compositions in this article have been shown in this series before. However, your author considers them as among the most interesting chess compositions, and as such worth of being shown again in this end of the year special.)

Links:

• Previous issues of the "Study of the Month"...