Poisoned pawn variations

Poisoned pawn variations


Till I've looked at the following games, when I was hearing for a poisoned pawn variation, only the one of the Sicilian Najdorf was coming up. And I've found namely five! Generally, looking at them, the common pattern is an unprotected Knight's pawn, being en prise by the Queen. However, only in one this pawn could be considered truly poisoned - toxic. In the others the opposite side can take it and retain an equal position.

They are presented briefly by order of appearance, mainly through the first found games played. But also trying to give some more modern aspect, where I can. And besides the second one where the poisoned pawn is really toxic, in the rest cases they are examined the variations where this pawn is taken.



King's Pawn Opening, Latvian gambit, Mayet, Poisoned Pawn variation [1.e4 e5 2.Nf3 f5 3.Bc4 fxe4 4.Nxe5 Qg5 5.d4]          up

  • Boudewijn Johannes van Trotsenburg vs Bernard Beffie, Amsterdam 1917

In this variation, black Q can take on g2 and it seems the best. While white lets it in shake of a better development. It's a really old opening approach, not really preferred, reminding a little the King's gambit setups. The name Latvian was given cause of Karlis Betins [or Behting] who analyzed this opening in the early 1900s.

Its origins concerned me a lot. And I can't say that I've managed to cover it 100%. Latvian gambit [1.e4 e5 2.Nf3 f5 3. Nxe5] is one of the first recorded openings. The idea seems appearing for first time in Damiano's book of 1512, but as opening where black has been given the odds of f pawn, with a main line of 1.e4 e5 2.Nf3 f5 3. Nxe5 Qh4+ [last move is checking cause of the odds].

From Damiano 1512

This gambit was first analyzed in a standard setup by Gioachino Greco, found surely in manuscripts & books since 1620. Greco was responding to 2...f5 with 3.Nxe5 or 3.exf5, although in one early MS I couldn't track this 2nd response. Pure Latvian gambit and called, cause of this analysis, for centuries as Greco's counter-gambit. As you can see they aren't any references yet of the poisoned pawn variation.

[I'm mentioning here the early checked by me works, used mainly for a following topic I was researching: four copies of italian manuscripts of the John G. White Collection in Cleveland Public Library, dated in 1623-1625: it.1 untitled work of 1624 , it.2 The Booke of Ordinary Games, 1763 copy of a 1623 italian MS , it.3 italian Trattato-Samuda MS of 1625 , it.4 italian Trattato-MS of 1624 , en.1 The Royall Game of Chesse-play, english edition of 1656 (Latvian gambit in Gambetts 39-42) , fr.1 Le jeu des eschets by Gioachino Greco, french edition of 1669 (Latvian gambit in livre 2, chapitres 40-43) , de.1 Das Schach des Herrn Gioachino Greco by Moses Hirschel, german edition of 1784 (Latvian gambit in spiels 40-43) , en.2 C. Greco by Lewis, english edition of 1819, (Latvian gambit in games 15th & 16th) , en.3 The games of Greco by Hoffman, english edition of 1900 (Latvian gambit in games 70-72) , misc.1 Greco and his manuscripts by John G. White, 1919].

front page of italian Trattato-MS of 1624 Latvian gambit in untitled work of 1624


Footnoting just to mention that this Latvian gambit variation, 1.e4 e5 2.Nf3 f5 3. Nxe5, is said to be found firstly in a game Polerio vs da Cutri. If it's a real game, it's dated probably before 1600. Trying to search for it anywhere, I've tracked the game's move sequence in a copy of a French 1827 copy of an earlier Florence 17th century MS, of J. G. White's collection. Murray, p. 821 (6), dates the original MS not before 1640, so actually after Greco. A more interesting thing is that the following written move sequence-order (n.23) is a touch towards the poisoned pawn variation: 1.e4 e5 2.Nf3 f5 3. Bc4 fxe4 4. Nxe5 d5 5. Qh5+ .... And possibly again from a Polerio vs da Cutri game.

copy of the Florence 17th century MS


After this we can find Carlo Cozio, in his Giuoco degli scacchi, vol1, 1766, p. 287, just suggesting a line: 1.e4 e5 2.Nf3 f5 3. Bc4 fxe4 4. Nxe5 d6... But the first clear poisoned pawn variation I've found is in Bilguer's Handbuch des Schachspiels by von H. v. d. Lasa, 1843, p. 57. Direct reference for the opening was Aaron Alexandre's Encyclopédie des échecs, 1837, but I couldn't find any copy of this. Lasa is generally mentioning many other previous sources for each opening - and I've checked them all! None had the poisoned pawn variation. So according to my findings it's the first clear reference.

Paul Rudolf von Bilguer Baron Tassilo von Heydebrand und der Lasa Aaron Alexandre



The problem:

During my search, I've tracked a game NN vs Gioachino Greco, Rome (1590), 0-1 in https://www.chessgames.com/perl/chessgame?gid=1481289! And it's a pure Latvian gambit poisoned pawn variation game!?!?! Of course I've felt many question marks hitting me! And reading at the comments of the game, I wasn't the only one. I don't know where this game came from - probably there's a source but it has been kept secret. As I've checked all the possible MSS and older versions of Greco's work and this game-opening there's nowhere, dated actually in 1590?!?

My questioning became really bigger as v. d. Lasa, in his really systematic work, reprinted-updated many times, doesn't mention any familiar source! Only one possibility I could think: there's a 1706 MS, the so called Caze MS [in Cleveland Public Library], that according to Murray, p. 822 (7), belongs incorrectly in the Polerio group, including inter alia some Greco's games!

Really don't know! Anyway...

So according to what I've found, the first played poisoned pawn variation was in Amsterdam of 1917...

from Tijdschrift van den Nederlandschen Schaakbond 6/1917
Dutch text Google translation a little modified

Kan iemand zich vreedzamer, rustiger menschen voorstellen dan de heeren Beffie en v. Trotsenburg? Zij spelen een match. Maar wie denkt hierbij aan vreedzame rustige remisepartijen?

Natuurlijk niemand! Leer om leer, is hun leus. Het staat nu 2—2.

Hier volgt de laatste partij, den 31en Maart gespeeld. Uit theoretisch oogpunt is zij belangrijk om den nieuwen zet 5. d2—d4. Of ze Greco's zet 2... f7—f5 weerlegt? Wie durft in dezen tijd van lichtbesparing eenig licht ontsteken? Behting, de Riga'sche expert zegt, dat op 5. d4 eenvoudig Dg2: volgt. Maar dat was nog voor de revolutie!


Let op, hoe pion d4 hier meewerkt.

Zaterdag 21 April deelde Dr. Beffie ons onder strikte geheimhouding mede, dat de variant niet deugde. Voor Wit of voor Zwart? Ja, dat is juist het geheim!

Can anyone imagine more peaceful, more quiet men than Mr. Beffie and Mr. Trotsenburg? They play a match. But who is thinking about peaceful, quiet draws?

Of course nobody! Leather for leather is their slogan. It is now 2—2.

Here is the last game, played on March 31st. From a theoretical point of view, it is important for the new move 5. d2-d4. Whether it refutes Greco's move 2 ... f7-f5? Who dares to turn on any light in this time of light economy? Behting, the expert from Riga, says that on 5.d4 simply follows Qxg2. But that was before the revolution!


Note how pawn d4 cooperates here.

On Saturday, April 21, Dr. Beffie informed us, under strict secrecy, that the variant was not good. For White or for Black? Yes, that's the secret!



Dr. Bernard Beffie in https://www.joodsmonument.nl/nl/page/26951/bernard-wolff-beffie


Queen's Pawn Opening, London system, Poisoned Pawn Variation [1. d4 d5 2. Nf3 Nf6 3. Bf4 c5 4. e3 Qb6 5. Nc3]          up

  • Vidmar vs Reti, Kosice 1918

Not a so popular variation. However, it's the only one where the poisoned white b-pawn is really a toxic one. So a puzzle is following with this trap. Black took the pawn with 5..Qxb2...


The first game, played with this variation, that I've found in databases, is the Vidmar vs Reti within Kosice 1918. Tried to confirm it in newspapers or chess literature, but in vain. The closest reference, by which a conclusion can be derived, is in Larry Evans' Vienna 1922, where the Maroczy - Vukovic game of the same variation is described. There is not any mention about previous games of it, so at least it isn't disproved. This maybe concerned me a little more than it should for this particular case, but it was an almost immediate reaction after looking for the origins of the Latvian variation and of the following 4th Najdorf variation. Just to say that it can be found at least in two online databases, but also in a download link of an informative article for this tournament [in zanchess]. Anyway...

Kosice [Kaschau in German, Kassa in English] is a city of Slovakia, at the time part of Austro-Hungarian Empire. Few months later, after the end of World War I, it was absorbed by Czechoslovakia. And this is a point that attracted some attention...

"It is surprising to learn, through the pages of the 'British Chess Magazine' and 'La Strategie', that an important masters’ tournament was held at Kassa, in Hungary, from August 5 to 19, almost on the very eve of the collapse of the Central Powers." [from American chess bulletin. v.16 (1919), p. 9].

It was the 1st Charousek Memorial and the first international success of Richard Reti. A major international chess event, taking in account the ongoing war, with strong participants such as Schlechter, Vidmar, Mieses. It was organized by the Kassa chess club during Aug 1918 [the names of Dr. Vecsey & Abonyi are mentioned around its set-up]. This Kassa chess club seems to be a Hungarian one, that ceased to exist after the war and was absorbed-substituted by a Czechoslovak one [sources: Gyula Breyer by Jimmy Adams, p. 464 & Jubilejný almanach mesta Košíc a východ. Slovenska: 1918-1928, p. 183-184].

from Tijdschrift van den Nederlandschen Schaakbond, 9/1918

It should have been fascinating! Vidmar in great shape, coming after a win in the four Masters tournament of Berlin 1918, and young talented Reti just starting to rising. They played on the last 11th round, while Vidmar needed only win to tie on top with Reti...


Schlechter vs Reti in Kosice 1918, Holubčík archive found in https://goout.net/en/exhibitions/richard-reti-sachovy-genius/uovve/+thmfo
Milan Vidmar in Karlsbad 1911, the closest I've found, in WSZ 1911, 304


French Defense, Winawer, Advance, Poisoned Pawn Variation [1. e4 e6 2. d4 d5 3. Nc3 Bb4 4. e5 c5 5. a3 Bxc3+ 6. bxc3 Ne7 7. Qg4 ...]          up

  • Bohatirchuk vs Alatortsev, Leningrad, 9th USSR Championship 1934/35

The variation is called a poisoned pawn after white's 7. Qg4, threatening to take on g7. However, this can't be correct dogmatically. But maybe there's no other way to name it. After white's 7. Qg4 the pawn can't be considered poisoned yet, as it's not certain if it will remain an easy target. Black can short castle or even play 7... Kf8 protecting it. It's black's 7th move that will give to the variation the poisoned pawn element I think. And this 7th possible move seems to be mainly 7... Qc7. But also eg 7...Nbc6 will let g pawn unprotected. The first appeared choice of 7... Qa5 probably doesn't allow black to let his g pawn taken.


And a trap for black that can occur two moves after the end of the opening of the previous pgn.


The opening idea should be attributed to Vladimir Alatortsev, who by black's ...Ne7 [firstly tracked since 1931] invited white to play 7. Qg4 and take on g7. But first that tried to take advantage of this unguarded black's g pawn was Vsevolod Rauzer on 4th round of 9th USSR Championship 1934/35. Rauzer [1908-1941] is known for his contributions in opening theory and has actually given his name to one of the Sicilian variations. Alatortsev's plan, playing in these first games ...Qa5, seemed to be letting his g pawn so to strike on c3 and generally on white's qside. However Rauzer's opening responses were pretty accurate, keeping the adv for white, and white lost in the end the game cause of middlegame inaccuracies. Fedir Bohatyrchuk, one of the most controversial figures of Soviet chess, dealt with the same plan on 13th round successfully.

Final results of 9th USSR Championship 1934/35 from Шахматы в СССР 1/1935
Vladimir Alatortsev Vsevolod Rauzer Fedir Bohatyrchuk



I don't know if it's because the opening has been worked out in depth, but looking at the games of the last two years, I can say that white has better results. Especially in higher levels. Two games of the last decade are following. One with 14. Ne2 and one with 14. Rb1



Sicilian Defense, Open, Najdorf, Poisoned Pawn Variation [1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 e6 7.f4 Qb6 ...]          up

  • Nezhmetdinov vs Sherbakov, Riga, USSR teamCh 1954

Whatever it was said about French Winawer I think it's valid here too. The variation is called poisoned pawn after black's 7...Qb6 threatening to take on white's b pawn. However white has at least one good choice protecting b-pawn, 8.Nb3. And 3 choices that leave b-pawn unprotected. Most popular by far 8. Qd2, the little dubious and criticized by theory 8.Qd3. But also 8.a3 that can create an easy trap that follows [some basics of the variation also in the pgn briefly]...


The origins of this variation concerned me a little.

Surely it was introduced internationally by Keres within Gothenburg Interzonal of 1955 in Sweden, where he tried as white at least twice this 8. Qd2 letting b-pawn en prise. Against Panno on round 12 who didn't take on b2, and vs Fuderer on round 16 who took and a brilliancy by Keres followed. Black's 7...Qb6 had already been seen played by Bronstein few months earlier within Belgrade 1954, but this 8.Qd2 was treated somehow as a novelty [eg in De waarheid of 15-09-1955, where the Keres-Fuderer game was presented under the title "Een Keresiaans pionoffer". Additionally and some months later in Leeuwarder courant of 24-02-1956, black's 7...Qb6 was criticised a posteriori to a degree suggesting instead 7...Qc7]. Looking at the annotations of both games in Tidskrift för Schack 4/1955, no solid conclusion can be derived. Comments are poor! Just to mention that Fuderer's 8...Qxb2 was marked with "?".

Gothenburg 1955 from Tidskrift för Schack 4/1955

However the variation seems that was played few months earlier. But by whom?

Two sources are mentioning a game Tolush - Furman, Leningrad, 1954, but without any reference of the prime source, or the exact month and event. One is Russian wiki article on Najdorf variation. The other is the Russian edition of Bobby Fischer by Nikolai Kalinichenko, p. 112. Tried to confirm it anywhere. In Russian Chess base I've tracked a Furman-Tolush game within Leningrad Championship of 1954 but it's of King's Indian setup. This game is mentioned in Abramov's Шахматы за 1954 год, p. 326, at least the first moves [the book can be downloaded in http://webchess.ru/ebook/438.html]. I couldn't find any other, but of course that's no solid proof. However I decided not to include it here.

In databases the earliest game I've found is Nezhmetdinov vs Sherbakov within USSR team championship of 1954 in Riga. Game that is confirmed in Abramov's Шахматы за 1954 год, p. 125, game #46 [it's noticeable that the book was sent for print in May 1955 according to front page credits]. So we have our game!

The 2nd USSR team championship [also mentioned as 2nd Soviet Team Chess Cup] was held in Riga during Sep 1954. Spartak team won. Nezhmetdinov of Spartak had the best individual results on 5th board and Sherbakov of Medik came 2nd. According to Abramov's comments at the end of the game, it seems that Sherbakov was a Candidate Master during the event [="Боевая партия! Молодой кандидат в мастера показал себя искусным шахматным бойцом." ~ few months later also Tal became a Master who participated in this event!]. After Sherbakov's poor results in the following USSR championship of 1955, he seems stopping playing at professional level.

Spartak team of 1954 from Шахматы в СССР 12/1954

The game, besides the opening novelty and an early intense moment, isn't so fascinating and I give it here more for the history reasons...


Two more recent games are following, based on popularity of the moves.

First with 9.Rb1 Qa3 10.e5. For this 10.e5, Garry Kasparov [commenting Tal-Tolush, 1956, in his My Great Predecessors, Part 2] has described it as: "An unprepared attack. Later on, recognising the drawbacks to the early e4-e5, White began playing 10 f5 or 10 Be2." Looking in the databases I can say that after some moves, it appears having some bigger draw percentage, while in the rest of the games black seems to have better results.


And one with 9.Rb1 Qa3 10.f5. White's most popular choice, creating a weakness on e6.


Trompowsky Attack, Poisoned Pawn variation [1.d4 Nf6 2.Bg5 c5 3.d5 Qb6 4.Nc3]

  • Vaganian vs Jansa, Kragujevac 1974          up

White lets b pawn to be taken, that it isn't really toxic. However black should afterwards retreat, or else there's trap. And generally looking at the results it seems that white has an advantage.


First games I've found in databases are two played by Rafael Vaganian as white in the end of 1974. And actually this variation is mentioned sometimes as Vaganian gambit. First one in Oct vs Vlastimil Jansa, within Kragujevac YUG 1974 that Vaganian actually won. 2nd two months later, within 42nd USSR Championship of 1974, vs Viktor Kupreichik. It seems that was a variation known in Leningrad at the time. In 64/49-1974 magazine, presenting the 2nd vs Kupreichik game, I've found in its comments the following...


Russian comment on 4...Qxb2 Google translation a little modified
Итак, гамбитиый вариант дебюта ферзевой пешки. Эту систему часто примеияют ленинградские шахматнсты. С одним из них - М. Цейтлиным - Р. Ваганян совершил недавно поездку в Югославию, в город Крагуевац. Из восьми побед одну ои одержал именно в этом дебюте. So, the gambit version of the queen pawn's opening. This system is often used by Leningrad chess players. With one of them - M. Tseitlin - R. Vaganian recently made a trip to Yugoslavia, to the city of Kragujevac. Of the eight victories, one won in this debut.


42nd USSR Championship 1974, front page of 64/49-1974 magazine. Vaganian should be 2nd from left, Kupreichik 2nd from right

So here the first vs Jansa game.


One that maybe shows a better opening approach for black, although lost.


And an uncommented fascinating one...


Which poisoned pawn seems more dangerous?

I think I would prefer being black on the French one offering the pawn and black on the Sicilian one accepting it!



Thanx for reading....