Open chess diary 41-60

The moves below are awesomely beautiful. Or ugly—hard to say. They're the longest "database endgame" mate, 262 moves. In 1991, Lewis Stiller already made the surprising discovery that this endgame, King plus Rook and Knight versus King plus two Knights (KRNKNN in databasese) is won for the strongest side in 78 % of the cases. He gave the longest win, which was 243 moves - but that was the distance to conversion (the reduction to a smaller endgame), not to mate. From that conversion to mate it was a further 3 moves; a total of 246 moves for the entire win. But for the fastest mate, you could not simply add those numbers, because Black could perhaps allow a quicker conversion to a slower mate, or White could perhaps allow a slower conversion to a faster mate. (See my story Stiller's Monsters on this site.)
It was expected that the direct path to mate, where both sides only care about the distance to mate, would be shorter than 246 moves. Surprisingly, it turned out to be longer: 262 moves. We owe this discovery to Ken Thompson, who constructed the (93 Gigabyte) database, and Peter Karrer, who found this longest mate in it.
Playing over these moves is an eerie experience. They are not human; a grandmaster does not understand them any better than someone who has learned chess yesterday. The knights jump, the kings orbit, the sun goes down, and every move is the truth. It's like being revealed the Meaning of Life, but it's in Estonian. On Thompson's Website, where this and other endgame databases can be found, he has named the link to them: 'Play Chess with God.'

White mates in 262
Ken Thompson, Peter Karrer, 2000

1.Kd6 N6g4 2.Kd5 Ne3+ 3.Kd4 Nc2+ 4.Kc3 Ne3 5.Kd2 Nfg4 6.Kd3 Kg2 7.Ke4 Nc4 8.Kf4 Nge5 9.Ra7 Nd3+ 10.Ke4 Nf2+ 11.Kd4 Nd6 12.Kd5 Nb5 13.Ra5 Nd1 14.Kc5 Nbc3 15.Kc4 Ne2 16.Kd3 Ng3 17.Ra1 Nf2+ 18.Ke3 Nf1+ 19.Kd4 Ng3 20.Ra3 Nf5+ 21.Kd5 Nd1 22.Nc6 Nde3+ 23.Ke6 Ng7+ 24.Kf7 Ngf5 25.Kf6 Kf2 26.Ra2+ Kf3 27.Ke5 Kg4 28.Ke4 Nc4 29.Ra4 Kg5 30.Nd8 Ng3+ 31.Kd4 Nb6 32.Ra7 Nf5+ 33.Ke4 Nc4 34.Ne6+ Kf6 35.Ra6 Ng3+ 36.Kd4 Nd6 37.Kd5 Nde4 38.Ra3 Nf5 39.Nf8 Nf2 40.Rf3 Ng4 41.Ke4 Nh6 42.Rf2 Ke7 43.Ng6+ Ke6 44.Nf4+ Kf6 45.Nd3 Ke6 46.Nc5+ Kd6 47.Nb7+ Kc7 48.Na5 Nd6+ 49.Kd5 Ng4 50.Rd2 Ne3+ 51.Kd4 Ng4 52.Rg2 Kb6 53.Nb3 Nb5+ 54.Ke4 Nd6+ 55.Kf4 Nh6 56.Rc2 Nb5 57.Rc1 Nf7 58.Nd2 Nc7 59.Nc4+ Kb5 60.Ke4 Ng5+ 61.Kf5 Nge6 62.Nd6+ Kb6 63.Nc8+ Kb5 64.Ke5 Nc5 65.Rg1 Nb3 66.Rg7 Na6 67.Rb7+ Kc4 68.Nd6+ Kc3 69.Rg7 Na5 70.Rg4 Nc6+ 71.Ke4 Na5 72.Ke3 Nb4 73.Nb5+ Kb3 74.Nd4+ Ka4 75.Rh4 Nc4+ 76.Ke4 Nd6+ 77.Ke5 Nc4+ 78.Ke6 Na3 79.Rh8 Nbc2 80.Nf5 Kb5 81.Kd5 Nb4+ 82.Ke4 Nc4 83.Nd4+ Kb6 84.Rh6+ Kc5 85.Rh5+ Kd6 86.Nf5+ Kd7 87.Rh7+ Kc6 88.Kd4 Kb5 89.Rh5 Na5 90.Nd6+ Kb6 91.Rg5 Kc7 92.Nf5 Kd7 93.Rg2 Ke6 94.Ng7+ Kd7 95.Rh2 Nb7 96.Rd2 Ke7 97.Rb2 Nd8 98.Kc4 Nbc6 99.Kd5 Kd7 100.Rd2 Ke7 101.Nh5 Nb4+ 102.Kc5 Na6+ 103.Kb6 Nb8 104.Nf4 Nf7 105.Kc7 Na6+ 106.Kc6 Nb8+ 107.Kd5 Nd7 108.Re2+ Kf6 109.Rf2 Nb6+ 110.Kc6 Nc4 111.Nd5+ Kg5 112.Ne7 Nh6 113.Kd5 Ne5 114.Rf1 Nd7 115.Rg1+ Kf4 116.Ng6+ Kf5 117.Kd6 Nf6 118.Nh4+ Ke4 119.Re1+ Kd3 120.Ke5 Nd7+ 121.Ke6 Nc5+ 122.Kd5 Nd7 123.Rd1+ Ke3 124.Rf1 Nb6+ 125.Ke6 Nc4 126.Ng2+ Ke2 127.Rf4 Nb2 128.Rb4 Nd1 129.Ke5 Nf7+ 130.Kf6 Nd8 131.Nf4+ Kf3 132.Nd5 Nc6 133.Rc4 Na5 134.Ra4 Nc6 135.Ke6 Nb2 136.Rh4 Nd3 137.Kf5 Ke2 138.Rc4 Nde5 139.Rc3 Kd2 140.Rh3 Nc4 141.Kf4 Nd4 142.Ke4 Ne2 143.Rh2 Kd1 144.Rf2 Ke1 145.Rg2 Kd1 146.Kd3 Nb2+ 147.Ke3 Nc4+ 148.Kf2 Ne5 149.Rg5 Nc4 150.Rg4 Na3 151.Kf3 Nc2 152.Ke4 Kd2 153.Rg2 Nd4 154.Nf4 Kd1 155.Nh5 Nc2 156.Rf2 Na3 157.Kd3 Nc1+ 158.Ke3 Nc2+ 159.Ke4 Ne2 160.Rf3 Nb4 161.Rh3 Nc1 162.Rh2 Ne2 163.Ng7 Nc3+ 164.Ke5 Nb1 165.Ne6 Nd2 166.Rh3 Nc6+ 167.Kd6 Nb4 168.Kc5 Na2 169.Kd4 Nc1 170.Rh7 Ke2 171.Rh2+ Kf3 172.Rh5 Ke2 173.Re5+ Kd1 174.Rf5 Ne2+ 175.Kd3 Nc1+ 176.Ke3 Nc4+ 177.Kd4 Nb2 178.Rg5 Nbd3 179.Rg7 Ke2 180.Ng5 Ne1 181.Rf7 Kd1 182.Ne4 Ne2+ 183.Kc4 Ng2 184.Rd7+ Ke1 185.Ra7 Kd1 186.Ra1+ Kc2 187.Ra2+ Kd1 188.Rd2+ Ke1 189.Rb2 Kd1 190.Ng5 Ne1 191.Rb1+ Nc1 192.Kc3 Nd3 193.Ne4 Ke2 194.Kd4 Kd1 195.Rb8 Ne1 196.Rd8 Ke2 197.Nc5 Kf3 198.Rf8+ Ke2 199.Rf7 Na2 200.Re7+ Kd1 201.Nb3 Nb4 202.Re4 Ng2 203.Na5 Nc2+ 204.Kd3 Nge3 205.Nb3 Nf1 206.Re8 Ne1+ 207.Kc3 Nf3 208.Nc5 Ng3 209.Kd3 Ne1+ 210.Ke3 Nc2+ 211.Kf2 Nf5 212.Rd8+ Kc1 213.Ke2 Ng3+ 214.Kf3 Nf5 215.Rd7 Nfd4+ 216.Ke4 Kb2 217.Kd3 Nc6 218.Rd6 Ne5+ 219.Ke4 Nc4 220.Rd3 Na5 221.Na4+ Ka2 222.Rc3 Na1 223.Rh3 N1b3 224.Kd3 Ka3 225.Rh4 Nc6 226.Kc3 Ncd4 227.Nb6 Ne2+ 228.Kc2 Ned4+ 229.Kd3 Kb4 230.Nc8 Kb5 231.Kc3 Kc6 232.Rh5 Kc7 233.Ne7 Kd6 234.Nd5 Kc5 235.Nb4+ Kd6 236.Rd5+ Kc7 237.Kc4 Kc8 238.Nd3 Kb7 239.Rd7+ Kc6 240.Ne5+ Kb6 241.Rf7 Ne6 242.Kxb3 Kc5 243.Rd7 Nd4+ 244.Kc3 Ne2+ 245.Kd3 Nf4+ 246.Ke4 Ne6 247.Rd5+ Kb4 248.Rd6 Nc5+ 249.Kd5 Na4 250.Kd4 Ka5 251.Rd7 Ka6 252.Nc4 Kb5 253.Ra7 Kc6 254.Rxa4 Kd7 255.Ra6 Ke8 256.Ra7 Kd8 257.Rb7 Kc8 258.Rg7 Kd8 259.Nb6 Ke8 260.Ke5 Kf8 261.Kf6 Ke8 262.Rg8 mate

When computer chess is discussed, I often think of what the Dutch writer Rudy Kousbroek said, some 30 years ago, after having visited a Sex Fair in Copenhagen: we should leave all those dildo's, vibrators and inflatable dolls to each other.
    A few years ago, Kasparov made a disgraceful mistake when he stooped to use an anti-computer style against Deep Blue. It was not a pretty sight to see our World Champion play 1.d3 and 4.a3, instead of playing chess and, if necessary, grandly losing with 6-0. A King fights on his own terms, not on those of some alien.
    When Kasparov lost anyway, I was sure we had that behind us. Hurrah, mankind had once more constructed a machine that can lift heavier stones than man can lift himself. But in Holland, where everything happens 50 years later, this insight has not sunk in yet. On April 5, the Dutch Chess Federation will announce that in the coming Dutch Chess Championship (7-19 May), a machine will compete. They will say that this will generate interest in chess. Last year they had a woman for that. She was nowhere near the qualifying norm but, being the first woman to ever compete in our national championship, she could be used to generate interest in chess.
    It's sad, a chess federation which thinks generating interest in chess cannot be left to chess.
    Why not have Boris Becker in the Dutch Chess Championship? He'd generate interest. He's not a Dutchman, but neither is Fritz. Why not a tulip? It can't play chess, but neither can Fritz.
    What a pity I don't play in that championship. I'd generate some extra interest by teaching them a lesson and resigning at move 1 against their machine.
    What if the machine becomes Dutch champion? Then of course it should play first board on our national team. But, being eminently clonable, it should also play second, third and fourth board. And if other nations follow us, we will have matches like:

Netherlands - Germany
Board 1: Fritz - Fritz
Board 2: Fritz - Fritz
Board 3: Fritz - Fritz
Board 4: Fritz - Fritz

Postscript 5 April: Hail to the Brave.
The players have agreed to play the machine, except GM Paul van der Sterren, who will now simply start the tournament with 0 out of 1.

When you're a chessplayer, people will often tell you that they know chessplayers, too.
    A Dutch friend, whose father was born a German, once told me he had, in 1965, looked up his father's brother who still lived in Germany. The two brothers had been born in 1911 and 1913; their mother had died in childbirth of the youngest and not much later, their father had fallen in the First World War. The boys had been raised by their very wealthy grandparents in a castle-like house, with a coach house, servants, a beautiful garden giving on a river. But it had also been a Prussian upbringing; they had to stand when they had dinner.
    Both brothers, as was the custom in their world, completed Gymnasium a as well as ß. The youngest, my friend's father, had gone to Holland before the Second World War, and had become a successful businessman here. My friend did not know much about his uncle who had remained in Germany—only that he had studied various subjects at University without obtaining a degree; that he owned a small mettalurgic factory the running of which he left to a manager; that he had been married to a ballet dancer with whom he had a son, and that he was supposed to be 'quite a chess player.'
    When my friend looked up this uncle in 1965 in Wuppertal, he found a bitter and lonely man in his middle fifties, living in a small one-room appartment, the sink full of empty food cans from which he apparently ate directly. On a little orange crate there was a chessboard, and all around on the floor there were piles of chess books and chess magazines. Of their conversation, my friend only recalls his uncle's grumbling about the stupid newspapers who called a stray whale which had been spotted in the Rhine, a 'white baluga' - when the word baluga already meant 'white whale.'
    My friend never met his uncle again, and, around ten years later, heard he had died.
    One of the worldwide legion of nameless chess addicts, I thought, who play their bad games in chess café's, and whom you will see in the demonstration halls at chess tournaments, asking stupid questions.
    Then, leafing in a booklet about chess in Germany in the early postwar years, I suddenly saw his name. As 'Meister X' no less, which is, for the sake of reverence, how I will call him. Just before and after the war, Uncle X had played in tournaments against the likes of Eliskases and Unzicker, had won games against Sämisch and threefold German champion Kieninger, had had results of around 50 % in a few minor national masters tournaments. In the first postwar championship of Germany, Weidenau 1947, he had been 11th of 20, with 9½ out of 19.
    He had not been 'quite a chess player', as ignorant family members will phrase it, but a real one, a 'master', somebody who had been willing to sacrifice ordinary life for the chess dream. He just lacked the talent.

Kieninger - Meister X, Krefeld 1946
1.e4 e5 2.f4 exf4 3.Nf3 Be7 4.Bc4 Nf6 5.d3 d5 6.exd5 Nxd5 7.Bxd5 Qxd5 8.Bxf4 Qc5 9.Nc3 O-O 10.Qd2 Be6 11.O-O-O Qa5 12.Rhe1 c6 13.Nd4 Bb4 14.Re5 b5 15.Rg5 Threatening Rxg7+, but Black's next move refutes this untimely agression. 15...Be7 16.Rg3 b4 17.Nxe6 fxe6 18.Qe3 Too much of a good thing; Ne4 might have still been OK for White. 18...bxc3 19.Qxe6+ Rf7 20.Rxg7+ Kxg7 21.Bh6+ Kg8 22.Rf1 (See diagram.)
22...Qd5 Or did he miss this? A backward guarding is always difficult to see. 23.Qc8+ After 23.Qg4+ there follows Bg5+ with a back rank mate. 23...Qd8 and White resigned, because after Qg4+ once again the countercheck Bg5+, follows, and White is mated on f1.
After all, more of an adventure gone bad by White than a beautiful game by Black.

In its first 2000 issue, the Dutch Probleemblad nominated 32 compositions, four in eight categories, for Problem (or Study) of the Millennium. There were classics (Loyd's Steinitz Gambit; Yarosh' Babson Task; the Saavedra; Forsberg's multiplets), and problems I'd never seen before. Here's a little selection.

Mate in 3
Sam Loyd
'The Steinitz Gambit'
1st Prize Checkmate 1903

Loyd's classic - one of the most famous problems of all time. I'm glad that a century later, experts still rank it this highly.
Even if you've seen it a hundred times, the solution is fresh. 1.Ke2!! and now:
1...Kxe4 2.Bd3+ Kd4 3.Rf4 mate
1...Kd4 2.Rf4+ e5 3.Nxg3 mate
1...f1N+ 2.Rf2+ Kxe4 3.d3/Bd3 mate
1...Nc1+ 2.Ke3 and mate
And the glorious main variation:
1...f1Q++ 2.Ke3!! and on Black's ten checks, White always has mate. On other moves, too.
Loyd composed this in something like 15 minutes, on the streetcars from his home to his office.

Mate in 4
G. Bouma
1st Prize Probleemblad, 1969

White has four potential mates with his knights (on c7, c5, f4 and g5) all of which are defended by the black queen, and two of which are defended each by the Ra5 and the Bh2. With the key, 1.f7, threatening 2.f8N mate, these defenses are brought in disarray:
1...Qa3 2.Nc5+ Qxc5 3.Ng5+ Qxg5 4.f8N mate
1...Rf5 2.Ng5+ Qxg5 3.Nc5+ Rxc5 4.f8N mate
1...Qf1 2.Nf4+ Qxf4 3.Nc7+ Qxc7 4.f8N mate
1...Bd6 2.Nc7+ Qxc7 3.Nf4+ Bxf4 4.f8N mate
Each time, the queen is lured to a square where it interferes with another defender, and thus becomes overburdened.

Mate in 6
V. Rudenko
1st Place WCCT 1975

After for instance 1.Rb1+ Ka3 White would have two mates, if they didn't each have a pair of defenses; Bd6 and Rxf8 against Bf8; Bxf3 and Rxf3 against Rxf3. With the key 1.g3!, White halves both pairs, and depending on which defense Black restores, he uses his Rb3 and Pa7 to first lure the other defender over a critical square, then cut it off there too:
1...Bxg3 2.Rb8+! Ka3 3.a8Q Bxa8 4.Rb3+ Ka4 5.Rb7+ Ka3 6.Rxf3 mate
or:
1...Rxg3 2.Rb7+! Ka3 3.a8Q Rxa8 4.Rb3+ Ka4 5.Rb8+ Ka3 6.Bf8 mate

Helpmate in 2
b)Ra6; c)Ba6; d)Na6; e)Pa6
H. Forsberg
1st Prize Pauly Memorial, 1935

In a helpmate, Black begins and helps White to mate him in the specified number of moves - legal ones only.
Here, the miracle is the five different pure mates born from so little material, each reached through a unique series of moves.
a) 1.Qf6 Nc5 2.Qb2 Ra4 mate
b) 1.Rb6 Rb1 2.Rb3 Ra1 mate
c) 1. Bc4 Ne1 2.Ba2 Nc2 mate
d) 1. Nc5 Nc1 2.Na4 Rb3 mate
e) 1. a5 Rb3+ 2.Ka4 Nc5 mate

Series selfmate in 42
B. Lindgren
1st prize Feenschach, 1989

White must make a solo series of 42 moves, forcing a selfmate with the 42th. All moves have to be 'legal' - the parentheses for the fact that moves are usually alternated. The theme will appear in red.
1.e8N 2.Nxf6 3.Nd5 4.f6 5.f7 6.f8Q 7.Qxf3 8.Qg2 9.f4 10.f5 11.f6 12.f7 13.f8B 14.Bxh6 15.Be3 16.h6 17.h7 18.h8R 19.Rxh4 20.Re4 21.h4 22.h5 23.h6 24.h7 25.h8Q 26.Qxb8 27.Qe5 28.b8B 29.Bxa7 30.Bc5 31.a7 32.a8R 33.Rxa4 34.Rb4 35.a4 36.a5 37.a6 38.a7 39.a8N 40.Nb6 41.Nc4 42.Qxe2+ Nxe2 mate
This exact series of moves is the only way it can be done - and it involves White promoting to all the pieces twice.

Proof game in 19.0
U. Heinonen
1st Prize Springaren, 1993
(correction)

Again; far from the bed of the everyday chessplayer. This is a 'precision game' - the idea being that 19 whole moves is the quickest way to reach this position, and that each of those moves is necessary - interchanges are impossible.
The solution is: 1. b4 c5 2. b5 Qc7 3. b6 Qg3 4. hxg3 h6 5. Rxh6 axb6 6. Rc6 Rxa2 7. Na3 Rxc2 8. Bb2 Rc4 9. Nc2 Rch4 10. e4 g6 11. Bc4 Bh6 12. Ne2 Be3 13. dxe3 e6 14. Qd3 Ne7 15. O-O-O O-O 16. Rxc8 Nbc6 17. Ra8 Rh8 18. Ra1 Ra8 19. Rh1 Nb8 and the amazing thing is that both the white rooks and the black rooks had to change places.

It's not often that my favorite theme happens in topflight chess. It did today, in Linares.

Anand - Khalifman, Linares 2000
1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Ba5 6.b4 cxd4 7.Qg4 Ne7 8.bxa5 dxc3 9.Qxg7 Rg8 10.Qxh7 Nbc6 11.f4 Qxa5 12.Nf3 Bd7 13.Rb1 O-O-O 14.Qd3 Nf5 15.Rg1 d4 16.g4 Nfe7 17.Rg3 Be8 18.h4 Nf5 19.Rg1 Ne3 20.Bxe3 dxe3 21.Qxe3 Ne7 22.Bc4 Bc6 23.Rb3 Kb8 24.Qxc3 Qc7 25.Be2 Nd5 26.Qc4 Qa5+ 27.Kf2 Ba4 28.Rb2 Rc8 29.Qd4 Rc3 30.Bd3 Rgc8 31.Rc1 Rd8 32.Rcb1 Bc6 33.Rb3 Rxb3 34.Rxb3 Nf6 35.Qb4 Nxg4+ 36.Kg3 Qd5 37.Bh7 Qd1 38.Kxg4 Rd2 39.Kg5 Bxf3 40.Bd3 Rg2+ 41.Kf6 (see diagram) a6 42.Rb1 and Black resigned.

Everybody knows Garry Kasparov changed his surname from Weinstein. Nobody knows he was Garry Bronstein first, and changed that to Weinstein. This I gather from an interview with his first trainer, Oleg Privorodsky, at the KasparovChess site.
Privorodsky says: "Garry came to the House of Young Pioneers and School Pupils at the age of seven. His mother brought him to us. It was almost funny when I asked him about his surname. He said Weinstein, but I thought he said Bronstein. I wrote him down as Bronstein and even told him that he had such a great “chess” surname because there was very famous grand master David Bronstein. Garry was shy and so he didn’t correct me. For a long time he took part in tournaments under the name of “Bronstein.” I seem to remember that after he told his mother about it she came to us at the tournament to rectify the error. I wasn’t near the tournament table, but when I returned I discovered that surname “Bronstein” was changed to “Weinstein.” It was then that we learned his correct surname. I find it very funny that for such a long time we all called him “Garry Bronstein” and he never corrected us."

54. 4 March: Unguarded Guard.

White to play

Looking at an old issue of the German magazine Randspringer, I saw a nice example of one of my favorite themes: the Unguarded Guard. By that I mean a line-check being defended by interposing something on an unguarded square.
The diagram on the left is from a game 'under tournament conditions' Ruf - Mephisto Polgar (computer), Germany 1993. There followed: 15.Qxh6 Qb6+ 16.Re3! and now 16...Qxe3+ 17.Kh1 and mate or 16...Qxb2 17.Qxh7+ and mate.

In his obituary of Lodewijk Prins in the recent issue of New in Chess, Hans Ree mentions first seeing Prins in the flesh in 1959, dressed up as Saint Nicholas, playing an alternating simul with a Black Peter impersonated by John Bink.
    Funny, I remember that too—like Ree, I was a participant in that simul. But the really memorable thing about it was Keres. He was present too, he was in Holland as one of the stars in the traditional grandmaster simul tour. And when one board was free, because Saint Nicholas had won that game, Keres jokingly sat down there, and asked the Saint if he could play, too.
    Great hilarity—imagine that!
    But that's where the joke stopped. I remember Keres' shy awkwardness, as it became clearer with each move that he was now stuck with his joke. They couldn't laugh each round at how funny it was. He couldn't just walk away.
    I don't remember how it ended, but in my mind the game lasted until well into the middlegame.
    Finally, somebody must have thought of proposing a draw.

Inspired by item 51 below, Arne Moll sent me this photograph by Lewis Carroll,

Black to play

from Lewis Carroll - Photographien, ed. Karl Steinorth, 1991. The caption reads: "The Misses Lutwidge: two of Carroll's five maternal aunts, playing chess, c. 1858."
Indeed, again a light corner in the right hand square. As far as Moll could reconstruct the position, it looks like the diagram on the left. He comments that if it's Black's move (which seems to be the case), 1...Re6 is the only move that doesn't lose. "But can White castle?" he wonders. "In problems and endgame studies he can, of course, but how about photographs?"

This made me think of another castling contribution Moll sent me some time ago. On the Internet server FICS, he once witnessed the following game:
theblob - nope (FICS 1997): 1.d4 Nf6 2.c4 e6 3.Nc3 g6 4.Nf3 Bg7 5.Bg5 O-O 6.e4 h6 7.Bh4 g5 8.Bg3 Nh5 9.Be2 Nxg3 10.hxg3 Nc6 11.e5 b6 12.a3 Bb7 13.Qc2 d5

White to play

14.Nxg5 Qxg5 15.cxd5 Nxd4 16.Qd3 Nb3 17.dxe6 Nxa1 18.f4 Qg6 19.Qxg6 fxg6 20.Bd1 Bxg2 21.Rh2 Bc6 22.b4 a6 23.Ra2 g5 24.Rxa1 gxf4 25.gxf4 Rxf4 26.Bb3 Kh8 (see diagram) 27.O-O-O As far as I know, this is the only case of an illegal castling with the other rook. Moreover, illegal moves are normally impossible on the chess servers, so this castling must be due to a bug in the software. 27...Bxe5 28.Bd5 Bxd5 29.Nxd5 Rd8 30.e7 Re8 31.Nxf4 Bxf4+ 32.Kc2 Rxe7 33.Rd8+ Kg7 34.Rd4 Be5 35.Rc4 Bf6 36.Rc6 h5 37.a4 h4 38.b5 axb5 39.axb5 and Black overstepped the time.

51. 10 February: I finally found one.
; Name, Width, Height ; Registration code (if you have it) ; Reglink opened in new frame?; Name of new frame for reglink ; resolution (1-8) ; Image 1 to load ; Image 2 to load ; Link 1 ; Link 2 ; Link 3 ; statusbar msg on image 1 ; statusbar msg on image 2 ; Flip effect img 1 (0 .. 7) ; Flip effect img 1 (0 .. 7) ; speed of fade (1-255) ; pause (value = milliseconds). ; Extra height (applet w. - img w) ; Flip curve (1 .. 10) ; Shading (0 .. 4) ; Red in background (0 .. 255) ; Green in background (0 .. 255) ; Blue in background (0 .. 255) ; Optional image over applet ; Over image X offset ; Over image Y offset ; Memory deallocation delay ; Task priority (1..10) ; Min. milliseconds/frame for sync Sorry, your browser doesn't support Java ; Msg in no java browsers
My neverending, but until now fruitless search in non-chess publications for chessboards with a light square in the right-hand corner, finally yielded one where they had it right. After careful study, I was able to reconstruct the position on the board too, which again reveals that these people knew what they were doing. Legal position; queen, bishop and pawn on logical squares. The only blemish is perhaps the girl's hand, which suggests she's going to play the Bc1.

There has recently been so much conjecture about the 'Keres - Botvinnik case' (Keres losing his first four games against Botvinnik in the 1948 World Championship Tournament, before winning the fifth when Botvinnik was already champion) that I asked Genna Sosonko what he thought about it.
    Sosonko knows the chessworld and the Soviet society very well (he lived there until 1972), and he was also the one who (together with Max Pam) asked Botvinnik the question, in 1991, that made the Patriarch say: "At a very high level, it was proposed that the other Soviet players would lose against me on purpose, in order to make sure there was going to be a Soviet World Champion. It was Stalin personally who proposed this." (See item 42 below.)
    A lot of people saw the smoking gun of foul play in that, and I concluded that item with: 'Botvinnik admitted to a high-level plan for foul play in the 1948 tournament—and that he and Keres were involved.'
    With that, I violated my own maxim never to even hint at plots unless they are the only way to explain something. Plan was too strong a word, and involved too suggestive of active involvement.
    As far as Sosonko knows the Soviet Union and the Soviet chess world, he strongly thinks that there was not an outright coercion in 1948 for Keres to lose his first four games against Botvinnik. But he also thinks there was a general atmosphere in which it was very clear to Keres that it would not be a good idea to beat Botvinnik too often, so to say.
    Sosonko thinks things might have happened like this. Stalin would have made an offhand remark like: "Comrade Botvinnik seems to be headed for the chess world title. That is very good." This might have been seen as a command and handed down as such a few levels, to a point were conceivably an undersecretary of Sports might have jokingly said something to Keres like: "You're not going to wipe the great hope of our nation off the board too harshly today, are you?"
    Sosonko stresses he sees this scenario as an outside possibility, something not to be dismissed out of hand - but nothing more. Probably, Botvinnik's remark in the interview would boil down to something in that vein having been said, or having been presumed by him to have been said. He was a very suspicious man, always seeing plots against himself.
    There were never any rumors about the '48 tournament in the Soviet chess world. But then, in the Soviet Union that Sosonko knew, it would have been entirely impossible for such rumors to have existed. Should one have suspected foul play in that tournament, "then these would have been thoughts one would only dare to think in a solidly locked toilet room."

Trying my skills in ChessBase at finding games on position, I looked for games where one side had three knights. It came up with this game that I'd never seen before.

Cohen - Meyer, US Open 1972
1.e4 e6 2.d3 d5 3.Nd2 Nf6 4.Ngf3 Be7 5.g3 O-O 6.Bg2 b6 7.O-O Bb7 8.Re1 c5 9.e5 Nfd7 10.Nf1 Nc6 11.h4 Re8 12.N1h2 f6 13.exf6 Bxf6 14.Ng5 Nf8 15.Qh5 g6 16.Qg4 Nd4 17.Qd1 e5 18.Ng4 Bg7 19.c3 Nf5 20.Qb3 Rc8 21.Bd2 c4 22.dxc4

White to play

dxc4 23.Qc2 Bxg2 24.Kxg2 Qd7 25.Ne3 Bh6 26.Ne4 Re6 27.Ng4 Bg7 28.Rad1 Qc6 29.Kg1 h6 30.Bc1 h5 31.Nh2 Ree8 32.Nf3 Nh7 33.Nfg5 Bh6 34.Nd6 Bxg5 35.hxg5 h4 36.Nxc8 hxg3 37.Nd6 Nh4 38.Qe4 Nf3+ 39.Kf1 g2+ 40.Ke2 g1N+ 41.Kf1 Qxe4 42.Rxe4 Rd8 43.Kg2 Nhxg5 (see diagram) 44.Bxg5 Nxg5 45.Rxe5 N1f3 46.Re8+ Rxe8 47.Nxe8 Ne5 48.f4 Ng4 49.Rd7 Ne6 50.Rxa7 Ne3+ 51.Kf3 Nd5 52.Nd6 Ndxf4 53.Nxc4 b5 54.Nd6 Nd3 55.b3 g5 56.Nxb5 Ne5+ 57.Ke4 and Black resigned.

Later today, I leafed through John Collins' My Seven Chess Prodigies (Fischer, Lombardy, Robert Byrne, Donald Byrne, Raymond Weinstein, Matera and Cohen), and suddenly saw a diagram with three black knights - this game.
Lew Cohen was 12 when he played it. He is certainly the least known of these prodigies. What happened to him? He just stopped?

In the penultimate round of the recent 'Young Masters' tournament in Groningen, leader Bokros from Hungary and his compatriot Horvath played a prearranged draw by which Bokros secured first place. As especially young players will sometimes do, they did not just play a few token moves, but copied an old and very wild draw.

Horvath - Bokros, 'Young Masters', Groningen 1999 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 Bc5 5.Nxf7 Bxf2+ 6.Kxf2 Nxe4+ 7.Kg1 Qh4 8.g3 Nxg3 9.hxg3 Qxg3+ 10.Kf1 Rf8 11.Qh5 d5 12.Bxd5 Nd4 13.Qh2 Qg4 14.Qxe5+ Be6 15.Bxe6 Qf3+ 16.Kg1 Ne2+ 17.Kh2 Qf2+ 18.Kh3 Qf3+ 19.Kh2 Qf2+ 20.Kh3 Qf3+ 21.Kh4 Qf2+ 22.Kh5 Rxf7 23.Bxf7++ Kxf7 24.Rh2 Qf3+

Black to play

25.Kh4 g5+ 26.Qxg5 Rg8 27.Qh5+ Qxh5+ 28.Kxh5 (see diagram) Ng3+ 29.Kh4 Nf5+ 30.Kh5 Ng3+ (28...h6 29.d4) 31.Kh4 draw
This had already been played during 1968 and 1969 as a correspondence game between the readers of the Russian children's magazine 'Pionerskaya Pravda' with White, and Mikhail Tal with Black. The only difference was that the pioneers omitted the repetition between move 18 and 20, and that from the diagram, the finish was: 26...Ng3+ 27.Kh6 Nf5+ 28.Kxh7 Rg7+ 29.Kh8 draw.
What Horvath and Bokros apparently did not know, was that in the diagram, Tal had missed an easy mate in six, as one little pioneer, Vadim Brodsky, later demonstrated: 29...Nf4+! 30.Kh6 (30.Kh4 h5 and Rg4 mate) Rg6+ 31.Kxh7 Rg7+ 29.Kh6 (29.Kh8 Ng6 mate) Kg8 and mate in 2 moves by Rg6.

46. 12 January: A Guided Tour of Chess

Today at The Chess Cafe, the start of my new series A Guided Tour of Chess.

45. 2 January 2000: Steinitz move

White to play

Steinitz (sim 29) - Enschede, Haarlem (Holland) 1896
1.e4 e6 2.d4 d5 3.Nc3 Nf6 4.e5 Nfd7 (see diagram) 5.Nb1 c5 6.c3 Nc6 7.Nf3 Be7 8.g3 O-O 9.Bg2 cxd4 10.cxd4 f6 11.exf6 Nxf6 12.O-O Bd7 13.Nc3 Rc8 14.Be3 Be8 15.Rc1 Bh5 16.Qb3 Bxf3 17.Bxf3 Na5 18.Qb5 a6 19.Qe2 Ne8 20.Bf4 Rc6 21.Bg4 Rf6 22.Bg5 Rg6 23.Bxe7 Qxe7 24.Nxd5 and Black resigned.

44. 27 December 1999: Whitaker - Capablanca

Recently, Sam Sloan published the game Whitaker-Capablanca, New York, 1913, saying: ''Whitaker nearly defeated Capablanca," and: ''Whitaker once had Capablanca completely busted," and: "Capablanca turned a losing position into a win." This is nonsense (Sloan doesn't even point out a moment where White is clearly better), and in the one line he gives, he misses the main tactics.

Whitaker - Capablanca, New York 1913
1.d4 Nf6 2.Nf3 d6 3.Nc3 d5 4.Bf4 e6 5.e3 Bb4 6.Bd3 c5 7.O-O Probably just an oversight; as a trap it seems to deep. 7...c4 8.Be2 Bxc3 9.bxc3 Ne4 10.Qe1 Qa5 11.Nd2 Nxc3 (Not Qxc3 12.Nxe4 Qxe1 13.Nd6+)

White to play

(see diagram)
    12.Bxc4 It must have been a great satisfaction in itself to be able to play such a move against Capablanca, but a bust? Far from it. Even after dxc4 Black has a forced draw if he wants it, but maybe Capablanca rejected the sacrifice on general grounds, or because he saw that after

White to play
After 14...Kd7 (analysis)

12...dxc4 13.Nxc4 (A reader, Allan Savage, mentions that with 13.Nb1, White might have a small edge - Ne4 works that way, too.) 13...Qb4 14.Nd6+ Kd7 (Sloan only gives Ke7 15.Ne4 Nd5 16.Bd6+ and wins) White can force a draw with (see the diagram on the right) 15.Nxf7 Re8 16.Bd6 Qb2 (or Qa5) 17.Ne5+ Kd8 18.Nf7+ etc. (Another reader, Gene Gnandt, shows that 18.f3! with the mating threat Qh4 wins. He gives 18...Nc6 19.Nc4 Qxc2 20.Rf2 Qd3 21.Nb2 etc. This in fact makes the following analysis of 15.Rb1 superfluous.)
    But White can also play for a win in the diagram: 15.Rb1! Qxb1 16.Qxc3 Qb6 17.Nxf7 and now Black has two

White to play
After 19...Kf6 (analysis)

possibilities, which I helped Rebel to analyze.
a) 17...Re8 Loses by force. 18.Ne5+ Ke7 19.Qa3+ Kf6 (see diagram) 20.e4! and now:
    a1) 20...Qxd4 21.Bg5+ Kxe5 22.Re1! and mate
    a2) 20...Nd7 21.Qg3 g6 22.Nxd7+ Bxd7 23.Be5+ Kf7 24.Qf4+ Kg8 25.Qf6 and White wins
    a3) 20...Nc6 21.Qg3 g6 22.Qh4+ Kg7 23.Nxc6 Qxc6 24.Be5+ Kg8 25.Qf6 and White wins.

White to play
After 25.f4 (analysis)

b) 17...Rf8 18.Ne5+ Kd8 19.Bg5+ Rf6 20.Nf7+ and now Black must go for a draw with Kd7 (21.Ne5+ Kd8 22.Nf7+ Kd7 etc.) because after Ke8 or Ke7 he is in great trouble: 21.Qxc8(+) Kxf7 22.Bxf6 Kxf6 (gxf6 23.Qh8 and White is better) 23.Qf8+ Kg6 24.Qe8+ Kf6 25.f4 (see diagram) with winning chances for White.
    Capablanca may have thought he was winning a pawn, but with 12...Nc6 he abandoned the idea. 13.Nb3 Qb4 14.Bd3 Na4 15.Qe2 Sloan: Qxb4 "would have secured at least a draw." Why? Because the position is then even? 15...O-O 16.Qh5 f5 17.g4 Qe7 18.gxf5 exf5 19.Kh1 Nb2 20.Be2 Sloan may be right that after 20.Rg1, White still has a good game. The rest is superior technique.
20...Nc4 21.Nc5 b6 22.Qf3 bxc5 23.Qxd5+ Be6 24.Qxc6 Rac8 25.Qg2 cxd4 26.exd4 Qd7 27.c3 Rf6 28.Rg1 Rg6 29.Qh3 Bd5+ 30.f3 Rg4 31.Rxg4 fxg4 32.Qg3 Qf5 33.Rg1 h5 34.h3 Rf8 35.Bxc4 Bxc4 36.Bd6 Rf6 37.Be5 Rg6 38.Kh2 Bd5 39.Qf4 Qc2+ 40.Rg2 Qd3 41.hxg4 Bxf3 42.Rd2 Qf1 43.g5 Qh1+ 44.Kg3 Bd5 45.Rf2 h4+ 46.Kg4 h3 47.Rb2 Be6+ 48.Kh5 Kh7 49.Re2 Qd1 50.Qd2 Qg1 51.Qf4 Bd5 52.Rd2 Bg2 53.Kh4 Kg8 54.Rb2 Kh7 55.c4 Qe1+ 56.Kg4 Qg1 57.d5 Bxd5+ 58.Qg3 Be6+ 59.Kf3 Qf1+ 60.Ke3 Qxc4 61.Bd4 Qc1+ 62.Rd2 Bc4 63.Qh4+ Kg8 64.Be5 Re6 65.Qd4 h2 66.Qd8+ Kh7 and White resigned.

PS 22 June: Well, not so superior, maybe. Gene Gnandt also showed that as late as move 63, Whitaker had a draw. (See Diagram) After 63.Qxh3+ Black must allow a repetition because Rh6 64.g6+! (Whitaker probably only saw 64.gxh6? Qa3+) Kxg6 65.Qg4+ loses

43. 25 December: Pogosyants' sleepless nights

Received "A (First) Century of Studies", in which John Roycroft presents a first batch of 100 endgame studies by the prolific Russian composer Erik (officially Ernest) Pogosyants, who died in 1990 at 55. He might well be the most prolific composer ever, with 6500 studies and problems, of which 4500 were published. Apart from that, he "penned some 4000 assorted poems and aphorisms, largely unpublished." According to Roycroft, Pogosyants was a rather naive man who, as a young and idealistic communist, criticised KGB-chief Shelepin and was subsequently arrested, sentenced, and treated in a psychiatric hospital. The substances administered to him there left him an insomniac for the rest of his life, which did not deter him from wearing his party insignas afterwards, and which was a blessing to the chess world. The quality of his work equals its quantity. Here are three early studies that occurred to him during his sleepless nights.

White to play and win
E. Pogosjants
1st Prize
Shakhmatnaya Moskva 1961 One of Pogosyants' first studies, and still one of his best. The passed pawn must be stopped. 1.Bf1 Bb5 2.Bg2 Bf1 Playing for stalemate, because after the forced 3.Bxf1 g2 4.Bxg2? would be just that. But with 4.Ng3! White can still win. After 4...Kxg3 5.Bxg2 or 4...gxf1Q 5.Nxf1 the Pf3 lives, and after 4...g1Q this pawn turns out to have been pinned against the guarding of h3, and 5.Nf5 mate is now possible.
There is the additional beauty that twice in this study (2...Bf1 en 4.Ng3) a piece is played to a square that has just been vacated by an enemy piece.

White to play and draw
E. Pogosjants
2nd Prize
Shakhmaty v SSSR 1963
In the position on the left, Black is a piece up and White has to make something out of his Pe6. But after 1.exd7? Be5+ and Bc7 Black wins, e.g. 2.Rf4 Bc7 3.Kg4 g6 Also losing is 1.g6 Be5+ 2.Rf4 (else the Rh7 escapes with a check) 2...dxe6
After 1.e7! Black must again try to move his rook from h7 with check, because after 1...Kf7 2.g6+ he would even lose. 1...Be5+ 2.Rf4! After 2.Kg2 Rh2+ or 2.Kf3 Rh3+, followed by Kf7, Black wins easily. 2...Bxf4+ 3.Kg4! (3.Kxf4 Rh4+ and Re4 or Kf7) Now 3...Kf7 is again followed by 4.g6+ and White wins. Therefore: 3...Rh4+ Vacating h7 to rob White of a future gxh7. After 4.Kxh4 Kf7 Black would have no more problems. But: 4.Kf5 Kf7 After 4...g6+ 5.Kxg6 White would even win. And now: 5.g6+ This study's central move. 5...Ke8 (or Kxe7) and it's stalemate.

White to play and draw
E. Pogosjants
1st Prize
Shakhmatnaya Moskva 1965

In the third study, the black pawn can't be stopped; White can only try to skewer the new black queen. But 1.Bb6 fails to Ke1 or Be5+ and most king moves are parried by 1...Bd4 and Black can interpose his bishop, e.g.: 1.Kf5 Bd4 2.Ba5+ Ke2 with promotion. White must therefore vacate f4 and guard e3 at the same time. 1.Ke4 Bg2+ is not possible, so 1.Kf3. Now Black must undo this guard. 1...Be2+ doesn't work: 2.Ke4 Bd3+ 3.Kf3 etc., and Bf4+ is still there. But: 1...Bg2+! 2.Kxg2 Ke3! Not Bd4 3.Kf3 Be3 4.Ba5+ Kd1 5.Ba4 draw. 3.Kg3 Bd2 4.Bf4+ Ke2 (Kd3 5.Bb5+ Kc3 6.Be5+ and Bb2) 5.Bf3+ It seems White is only giving spite checks. 5...Ke1 Because what to do next? Even here, I'm not sure I would find the fantastic idea that ensures the draw. 6.Bd6!! The threat Ba3 now forces promotion: 6...c1Q 7.Kg2! Threatening Bg3 mate, so Black must vacate d2. 7...Be3 If he uses the other diagonal, Bg3+ and Bf4+ cost him his queen. But now: 8.Bb4+ Bd2 9.Bd6 repeating the moves, and drawing. The bishop had to go to d6 on the sixth move; on c7 it would be attacked, and from b8 or e5 it wouldn't have had a (good) square for the saving check.

The book costs $ 9.95 and can be ordered at The Chess Cafe.

42. 10 December: Saying no to Stalin—the 1948 Botvinnik - Keres case

There have always been dark rumours around the World Championship Tournament of 1948: Keres lost his first four games against Botvinnik after weak play, and won the fifth when Botvinnik was already champion.
    In 1948, Botvinnik was the First Soviet World Chess Champion the nation was waiting for; a Russian, a model communist, while Keres was in a difficult position. He was an Estonian, had thereby become a Soviet citizen only in 1940, but a German subject after the occupation in 1941. Up to 1943 he played tournaments in Germany, after which he went to Spain, Finland and Sweden and, when the war was over, to Estonia which had become part of the Soviet Union again in the meantime. He only did so after having been promised that his playing in Nazi-Germany would be forgiven, but he had to promise he would leave his rights to a title match against Alekhine, which he had obtained before the war, to Botvinnik. He also was barred from playing important tournaments for a few years.
    When, after Alekhine had died in 1946, there was going to be a tournament for the world title in 1948, Keres could not be passed over. His 1-4 against the intended Soviet hero Botvinnik has always aroused the suspicion that the authorities had forced him to throw these games, whether Botvinnik knew about this or not. For more details, see two interesting articles at The Chess Cafe by Taylor Kingston; kb1.txt and kb2.txt. Kingston's conclusion is that, at this moment, there is no proof of foul play.
    However, there is an interview Max Pam and Genna Sosonko made in 1991 with Botvinnik, and which you can read in full (in Dutch) at Max Pam's site. When they asked the Patriarch if he had ever encountered collusion between Soviet players in his playing days, his answer was: "I have experienced myself that orders were given. In 1948 I played with Keres, Smyslov, Reshevsky and Euwe for the world title. After the first half of the tournament, which took place in the Netherlands, it was clear that I was going to be the new World Champion. During the second half in Moscow something unpleasant happened. At a very high level, it was proposed that the other Soviet players would lose against me on purpose, in order to make sure there was going to be a Soviet World Champion. It was Stalin personally who proposed this. But of course I refused! It was an intrigue against me, to belittle me. A ridiculous proposal, only made to put down the future World Champion. In some circles, people preferred Keres to be World Champion. It was disgraceful, because I had already proven by and large that I was stronger at that time than Keres and Smyslov."
    A nonsensical answer, but telling all the same. People preferring Keres? Botvinnik had been an intended Soviet Hero for at least ten years; the Estonian Keres with his tainted war past as the first Soviet World Chess Champion was unthinkable. And saying no to a proposal from Stalin? Even so: Botvinnik admitted to a high-level plan for foul play in the 1948 tournament—and that he and Keres were involved.

PS 12 December: Reader Roman Parparov had an interesting remark about this: "There is a reason why people would have preferred Keres to Botvinnik as the World Champion. It is not a reason you would think of as an intellectual and a civilized modern European. But this is quite an obvious reason to me. The reason is antisemitism. People didn't want the Jew Botvinnik to win the title. I was told by an elder friend of mine, that during the Botvinnik vs. Smyslov matches, people were becoming fan of Smyslov under the slogan of 'Vasiliy, kick the Jewish ass.' The antisemitic wave was very strong in Russia in 1947-53."

PS 4 February 2000: also see item 50.

41. 8 december: Most overestimated move of the millennium

Every time I see the diagram below, I feel the urge to commit sacrilege. The move White played there has been called brilliant, splendid, profound, it receives an average of 2 exclamation marks, and the last time I saw it, in Tim Harding's column The Kibitzer at the Chess Cafe, the game was one

White to play
Botvinnik - Capablanca
AVRO Tournament
Netherlands, 1938

of the Ten Influential Games of the Millennium, and the combination a 'great moment of inspiration'.
I'm sorry, but 30.Ba3 is the first move that comes to mind here, and it's pretty easy to calculate. First thing you notice is the blockading black queen, so you think of chasing it away with your otherwise dead bishop. Ba3 would also chase the queen from guarding Nf6—never know how that might come in handy. Something which also comes to mind is giving a check. You don't have to be a very strong player to hit upon the idea of Nh5+ followed by Qg5+ when you could take the knight with check if it were not defended. These two obvious thoughts combined lead to 30.Ba3 Qxa3 (we'll think about Qe8 when he plays that) 31.Nh5+ gxh5 32.Qg5+ Kf8 33.Qxf6+ Kg8 and now all we have to do is to check for a perpetual, either after 34.Qf7+ Kh8 35.e7, or after 34.e7 immediately. This is the hard part, but any 2200 could do it in 10 minutes. Let's start with 34.e7, it keeps the queen closer. 34...Qc1+ 35.Kf2 Qd2+ (Qc2+ just limits Black) 36.Kg3 Qd3+ (Qxc3+ 37.Kh4 Qe1+ 38.g3 or Qe1+ 37.Kh3 Qe3+ 38.g3) 37.Kh4 Qe4+ 38.Kxh5 Qe2+ 39.Kh4 Qe4+ 40.g4 Qe1+ 41.Kh5 game over; we don't have to worry about 34.Qf7+.
This was how it went (with 35...Qc2+) and where Capablanca resigned.