Nihal's problem: how the b-pawn mates

by Frederic Friedel
5/11/2017 – When the 12-year-old Nihal Sarin visited ChessBase in Hamburg in April it was not just his incredible chess skills and understanding that impressed us all. The lad is blessed with a fair deal of humour, and sprouts entertaining chess puzzles at every turn. Like the series of problems that took advantage of an imprecise definition of the castling rules. Or one, which we passed on to our readers: "A game ends on move four with a b-pawn giving mate." We got a lot of feedback, and the winner has been drawn: the prize, a Fritz 15 program signed by three World Champions, goes to...

ChessBase 17 - Mega package - Edition 2024 ChessBase 17 - Mega package - Edition 2024

It is the program of choice for anyone who loves the game and wants to know more about it. Start your personal success story with ChessBase and enjoy the game even more.


The Nihal Sarin problem

In April the 12-year-old super-talent Nihal Sarin (who now has one GM norm under his belt) visited us in Hamburg and entertained us with a lot of unusual problems. One of them we handed over to our readers: after defining the correct promotion rule as "when a pawn reaches the final rank you have to promote it, to any piece except for a king, on the same move," he gave us a final problem:

White to play and mate in two

André Schulz, who runs our German language news page, works for ChessBase Magazine and looks after our DVD authors, was thoroughly duped by this problem. He had been taken through the previous problems by Nihal, and decided that the last definition was waterproof. The picture at the top of the page shows André being given the correct solution by the lad: 1.Qb8+ Rxb8 2.axb8=bR#White promotes the pawn to a black rook! Where did we rule that out?

Incidentally in the FIDE handbook "Laws of Chess" we find the promotion rule impeccably formulated:

3.7 e: When a player, having the move, plays a pawn to the rank furthest from its starting position, he must exchange that pawn as part of the same move for a new queen, rook, bishop or knight of the same colour on the intended square of arrival.

The handbook even specifies how the move can be executed:

4.6: The act of promotion may be performed in various ways:

  1. the pawn does not have to be placed on the square of arrival,
  2. removing the pawn and putting the new piece on the square of promotion may occur in any order.

If an opponent’s piece stands on the square of promotion, it must be captured.

A b-pawn mates on move four

The second Nihal problem was one we had given to Vladimir Kramnik on Skype, and the former World Champion had found the solution in 33 minutes – while "taking a couple of calls in between".


Here the only moves that fulfil the requirement: 1.d4 c6 2.Kd2 Qa5+ 3.Kd3 Qa3+ 4.Kc4 b5#. You can enter them on the above board to fully enjoy the solution.

In our article we had asked our readers to send solutions and promised to draw a winner from all submissions. The prize: a Fritz 15 program, signed by three World Champions! They are Vladmir Kramnik, Anatoly Karpov and Viswanathan Anand. Before we announce the winner here some comments:

Hector N Salas Olaguer wrote: I immediately realized that is White which has to be mated, since this gives an extra half move while still satisfying the mate in the fourth move. What caused me some trouble is the interpretation of what a b-pawn means. Could be a pawn that by capturing could become a b-pawn, or does the original b-pawn mate but on a different file? Fortunately in this case the original b-pawn mates and still is in the b-file.

Rolf-Dietrich Beran of Altlandsberg, Germany, told us it took him all in all three houres to solve the puzzle.

Joshua Green wrote: "I actually found the problem much easier to solve than the similar problems you've posed before. I figured there was no reason not to let Black deliver the mate, thereby gaining an extra ply to work with. There's no time to promote a pawn, nor does there seem to be any way to support one that reaches the second rank. Mate by moving to the third rank seemed plausible, but where to put the white king? Support would still be required, and White naturally has a lot of protection along that rank. What about mate by moving to the fourth rank? Discovered mates can be considered – would they be allowed? But simple contact mates still seemed more likely. White king on c3 with a Black pawn on b4 is close, but there doesn't seem to be any way to cover all the light squares in time. At last, I considered mates by moving a pawn to the fifth rank. b5# was awfully tempting since it can be guarded in a few ways and the dark squares could easily be covered. The third rank had to be dealt with, but the black queen could take care of that. At first I tried bringing her out on the kingside, but when that was never fast enough I quickly hit upon her correct route."

Noam D. Elkies, mathematician and chess master: Although I haven't played a tournament game in 25 years, I found both solutions quite quickly – the r-promotion almost instantaneously, and the four-move helpgame in under five minutes. But then I've seen quite a few tricks with off-color promotions (even contributing one or two to Hochberg's Chess Braintwisters and helpgame tasks, including in a Benko column decades ago. Thanks for the puzzles.

David Rivas wrote: "I have enjoyed the article a lot, thank all of you. I have to say that I am going to use the b-pawn mate for a contest with my students. Thanks for this too."

Alberto Muniz Pardino, a Spanish chess trainer living in Hong Kong, wrote: "This problem took me more than one hour. I started wondering if after things like b5-bxc4 this still would be considered a b-pawn. I gave up soon as I don’t have enough white moves to get the king out and allow captures at the same time. So I started thinking about extraction of the king to c3 or c4. It took longer as I kinda assumed that the perfect square for the black queen was b2. This combines well with White playing d3 followed up by the king walk (so d3 square is not available). Unfortunately I didn't manage to convert my “illegal solution” into a legal one. 1.d3 e6 2.Kd2 Qf6 3.Kc3?!?! Qxb2?!?!? 4.Kc4 mate. It took me a while to find the similar pattern with the queen on a3 and the white pawn on d4. Thank you very much, it was fun!"

Marten Rosselli wrote: "It took me about five hours to find the solution (ten times more than Kramnik...). The idea, that it must be Black who mates, was quite obvious to me (you "win" half a tempo to meet the requirements of the puzzle), but I still had no idea how to construct it. I was looking at Ba6 combined with b4# solutions (a discovered check, instead of the pawn giving check directly – reading the rule discussion before, I thought this could be the neat trick that people won't think about). But those variations never worked out, so I focused on a solution with the pawn giving checkmate directly in combination with a black queen on the third rank covering important escape squares. Using other pieces to cover the escape squares would take too many tempi. However, I was trying to get the queen to the third rank via the kingside (via 1...e5 and the d8-h4 diagonal) which is simply too slow (takes one more move to get the queen to the queenside once again). Once I started trying to get my queen to the third rank via the queenside, I found the solution. First thinking about the possible mate position and then constructing the move sequence in a retrospective way was helpful too. Very nice article!"

Mark van der Hoorn: "The puzzle is nice, but not as difficult as the NxR# one was. The b3 square is the problem, but there are only limited ways to solve that. And once you do the solution is reasonably straightforward. Cheers for these. I like these sort of problems."

Lubomir Kovac wrote: "The b-pawn was really a challenging puzzle. I thought first about funny solutions. In the statement there is nothing written about a classical chess game – so it is possible to find a game in Chess960 which leads to mate on 4th move with b-pawn giving the mate. I found couple of them. But then I remembered that Kramnik solved it in 33 minutes. I started thinking that probably he didn't try to evaluate positions in Chess960, so it must be a normal start position. Anyway thanks for all the puzzles – I never thought about bending rules like this – especially the promotion rule."

Valentin Marcel wrote: "Nice one! I remember working on the knight takes rook mate in five when I was in college with my friends. That was really nice. Sadly I can't remember the solution, maybe it was pawn takes rook with knight promotion, can't make it work for the moment... But thank you again! Au plaisir de vous lire, Frederic."

Jochen Schoof: "Thanks for the entertaining time spent on these puzzles."

Nikolaos Alimpinisis from Greece wrote: "I found the solution in less than Kramnink's 33 minutes (my word for it!). After finding the idea of White's king going out to the c-file to be voluntarily executed by Black's b-pawn, I started looking for Black's possible moves controlling the squares around the king. The most suitable piece for this is the black queen, and that gave me the inspiration for the (helpmate) solution. I hope I will be lucky for the Fritz program. Thanks for these great puzzles!"

Jean-Philippe Karr: "I was happy to be quicker than Vladimir Kramnik... albeit without answering any phone calls. Thanks for this nice article."

Alejandro Ovejero: It took me a while to get the idea that the black b-pawn should be the one that mated, and not the white one. I think I got it in less time than Vladimir (OMG!!!). Congratulations for the great columns and coverage. I'm normally a silent viewer, but this time I take this opportunity to thank you for your dedication and passion in making this website."

Hector Salas Olaguer, Mayaguez: "Dear Fred, delightful article, great guest."

Tim Wee from Australia supplied the wrong solution (1.d3 c5 2.Kd2 d5 3.Kc3 Qa5+ 4.b4 Qxb4#), which he worked out during his honeymoon in Italy. It was accompanied by a nice little story which we share with you:

Tale Of The Treacherous Pawn

There once was this pawn, let's called him Bee. He was a lowly soldier who served the White Monarch. However, having previously been convicted of insubordination, he was relegated to the sidelines. And just to poke fun at his name, he was ordered to man the b-file, where he never really got the chance to contribute to any battle. Instead, he had to do annoying and menial tasks, like preparing the bishop for his usual "fianchetto" activities, something Bee never really understood.

But things got from bad to worse. In the last battle against the feared Black Monarch in the region of Sicily, the White Monarch was convinced by his evil Queen to sacrifice Bee in order to gain the upper hand against his enemy. Bee survived after the pragmatic Black Queen decided against capturing Bee, calling him a "poisoned pawn". But Bee decided that he had had enough. "That's it, I'm going to take down this White Monarch even if it'll cost me my life!", he exclaimed.

So, that very night, Bee poisoned his King with a hallucinogen. The King, in delirium, ordered the d-pawn guarding his chambers away and marched to the battlefield location designated C3, where he imagined his beautiful queen was waiting (she was in fact fast asleep along with the rest of the White Army). Now the alert Black army had noticed this unusual movement, and the Black queen advanced to A5, where she had the White King cornered. "Surrender, or die!", she demanded. But the White King had fallen asleep at C3 and wouldn't move (to B3) or respond at all. Now, under the strict battle treaty all countries had signed ("The FIDE Laws"), the Black Queen had to wait for an action from White (either a move or a resignation), so things were at a stand-still. And here is when the treacherous b-pawn, jumping to the B4 position, yells at the Black Queen "Capture me!" Black Queen duly captured the b-pawn giving mate, and the game (battle) ended with a b-pawn giving mate.

No, Tim, it is the pawn itself that must deliver mate. Bad solution, good story.

And the winner is...

Roland Kensdale from Aberdeenshire, UK. In his submission Roland wrote: "Nihal's Chess pranks are an interesting idea to remind people of how chess laws (here pawn promotion) developed. Light hearted and not too difficult problems in article. Entertaining. One of the comments in feedback said Chessbase had too many reports of young (Indian) players. Probably many older readers will feel a little jealousy, a common state of affairs in chess where younger players are getting stronger and stronger. I am not sure there would be a great appetite for articles on average ability middle aged players with declining powers."

Editor-in-Chief emeritus of the ChessBase News page. Studied Philosophy and Linguistics at the University of Hamburg and Oxford, graduating with a thesis on speech act theory and moral language. He started a university career but switched to science journalism, producing documentaries for German TV. In 1986 he co-founded ChessBase.


Rules for reader comments


Not registered yet? Register