All the 
best 
in chess.
(TM)

  FM A. Gunnarsson (2289) - GM I. Sokolov (2658)  
[A21]
 ICT / Jonsson Memorial (open) 
  Reykjavik, Iceland; (R9) / 23,10,2001.  

  [A.J. Goldsby I]  

I think I first saw this game shortly after it was played. (Probably in a magazine - or in the Informant.)  

A student and I were studying a book on the English, (in January, 2006 - see the bibliography at the end of the game); and we went over this contest again. I was so impressed with the tactics towards the end of this struggle, that I decided to briefly annotate this clash ... 
and I also featured it as my "Problem of The Week." 

It IS an interesting game ... and definitely deserves to be better known than it actually is.  

 1.c4 e5;  (Pawn formation.)  
The one and only Mikhail Botvinnik, {a former World Champion}; did not approve of this move in the English. I guess his logic ran something like: "If the Sicilian is good for Black, 
then a reversed Sicilian, (for White); must be even better!"  {w/ extra tempo}  

Despite this, 1...e5; remains a favorite choice of many GM's - even today - and is played at 
virtually every level of play.  

 2.Nc3 Bb4!?;  (tempo loss?)   
A rather eclectic move at best. (But since it is played by many GM's, apparently even strong 
players do not feel that it is completely unplayable.) 

     [ More 'normal' is the following continuation:   
       (>/=)  2...Nf6;  3.Nf3 Nc6;  4.g3 Bb4!?; "~"  ("+/=")   
       when White will gain a small edge, even after correct play by both sides.  

       [ See any good reference work, like MCO-14,  page # 666.  
         Columns one through five, and also all associated notes. ]   

        A good, recent example would be:  GM E. Bareev - GM S. Rublevsky;  
        / The Tournament Championships (R9) / Sochi, RUS; 2005. 
        {Drawn ... 1/2-1/2, in 99 moves!} ]   

 3.Nd5! Bc5;   
This move looks eminently logical, but the Fritz9 book gives it a full question mark. ('?')  
However, I find it extremely difficult, (impossible, really); to agree with this particular assessment.  

     [ Black could also play: 
        3...Be7;  4.d4 d6;  5.Nf3 e4;  6.Nd2 f5;  7.e3 Nf6;  8.Nxe7 Qxe7;   
        9.d5 0-0;  10.Be2 Nbd7;  "~"   when the second player appears   
        to have nearly equalized here.  

        GM Y. Seirawan - GM A. Shirov; Biel Izt / 1993.  (0-1, 72 moves.)  ]   

 4..Nf3 e4!?;  hmmm  (Maybe - '?!')  {See the diagram - just below.}  
It looks a little early for Black to be trying this. Raetsky and Chetverik duly award this the dubious appellation, ('?!'); yet it is not 100% conclusive that this move is bad. (At least, not from the way that this opening was played ... but maybe White could improve upon his play.) 

  ***********************************************************  

  rnbqk1nr/pppp1ppp/8/2bN4/2P1p3/5N2/PP1PPPPP/R1BQKB1R w   

  ***********************************************************  

I am not saying that this move is not less than best ... in fact, it probably is! But judging from the computer's evaluations, I cannot prove that this move is inferior  ...not beyond all reasonable doubt.  

     [ Apparently, Black should instead play the continuation of:   
       >/=  4...c6; ('!')  5.Nc3 d6;  6.g3 Nf6;  7.d3 0-0;  8.Bg2 h6;  9.0-0 Be6;   
       10.b3 Bb6;  11.e3 Re8;  12.Ba3!? Bc7;  13.Qc2 Bf5;  14.h3 Nbd7;  
       15.Rad1 Rc8;  16.Nd2,  "~"  (Maybe "=/+")  {Diagram?}   

       Black equalized out of the opening, and then went on to win a long game.   

       GM Mikhail Gurevich (2688) - GM Zoltan Gyimesi (2571);   
       2nd EU-ch / Ohrid, Macedonia; (R #5) / 05,06,2001.   
       (This game is annotated in the CB 2005 "MEGA" database,   
         if you happen to have it.) ]   

 5.d4!?,  (center + tempo)  
A good move, and one that is favored by masters, at least according to the statistics of the 
(MEGA) database.  

     [ The try of:  5.Ng5!?,  (Maybe - '!')  "+/="  was worth serious investigation. ]  

This position has only occurred about a dozen times, according to repeated searches of various {different} databases.  

Normally, Black has played 5...Bf8; but this strikes me as retarding the second player's overall development. {Both players continue the opening phase by continuing to bring out 
new pieces, and also both parties will have castled before move 15.}  

 5...Be7; ('!')  6.Nd2 c6;  7.Nxe7 Qxe7;  8.Qc2 Nf6;  9.Nb3 0-0;  
This is OK, but maybe a bit too routine? (Maybe 9...h6!?)  

     [ (>/=)  9...d5!?;  10.Bg5 dxc4;  "<=>" ]   

 10.Bg5 Re8;  11.e3 d6;   
A cautious move, and this might be Black's best course of action.  
(Since White has two Bishops, it is not in Black's best interests to open up the game.)   

     [ Also possible was:  11...d5!?;  12.a3 dxc4!?;   This looks interesting. 
       (Playing the Queen to d6, or even ...Nbd7; were worth a look as well.)  

       13.Bxc4 Nbd7;  14.Rc1,  "+/="    when White has a small edge, but   
        Black's position appears to be solid and playable. {There were no   
        matches in any database for this position.} ]   

 12.Be2 Nbd7;  13.0-0 h6;  14.Bh4, "+/="  14...Nf8;  (re-deployment)    
White remains a bit better, he has the two Bishops and Black's position remains congested. 
 {See the diagram - just below.} 

  ***********************************************************  

  r1b1rnk1/pp2qpp1/2pp1n1p/8/2PPp2B/1N2P3/PPQ1BPPP/R4RK1 w  

  ***********************************************************  

I am not exactly sure where Black went wrong, (maybe move two?); but Black's game here - while reasonable - is not exactly what I would call ... "the ideal position."  

     [ Possible was:  14...b6!?;  {Diag?}   as one student ... 
       at our local chess club ... suggested. ]   

White to play, what's the best move that Gunnarsson has in this position?   
 15.d5!?,   (Maybe - '!')   
This fixes the Black Pawn on d6, making it a permanent target for attack.  

Shredder 9.0 prefers   >/= 15.Na5, "+/="  here for White. Another plan was 15.Nd2, followed by massing the Rooks in the center. When the time was right, White could then break with f2-f3. 

     [ Also interesting was:  15.c5 d5;  16.Na5, "+/="  with an edge for White. ]  

 15...c5!?;  "~"  (key position)   {See the diagram - just below.}  
Trying to close lines.   

  ***********************************************************  

  r1b1rnk1/pp2qpp1/3p1n1p/2pP4/2P1p2B/1N2P3/PPQ1BPPP/R4RK1 w  

  ***********************************************************  

An interesting situation has arisen, on the chess board and one that is rich in chances for both sides. 

     [ The authors of the English book recommend that Black play:   
       (>/=)  15...cxd5;  16.cxd5 Ng6;  17.Bxf6 Qxf6;  18.f4 exf3;   
       19.Rxf3, "~"  (Maybe "+/=")   {Diagram?}   
       with a slight pull. (White is supposed to have a small advantage   
       here, but Black looks perfectly OK to me.)   

   *********************************************************************************   

        Maybe the best line was:   >/= 15...Ng6!;  16.Bg3 c5; "~"  
        (Very close to equal.) ]  

 16.Na5 Ng6;  17.Bg3 h5;  hmmm  {See the diagram - just below.}   
Again - the authors of the English book brand this as dubious, ('?!'); and claim that 17...Bg4; was just a little better. And while they may be correct, the game might have wound up as a dull draw ... that no one would have remembered.   

  ***********************************************************  

  r1b1r1k1/pp2qpp1/3p1nn1/N1pP3p/2P1p3/4P1B1/PPQ1BPPP/R4RK1 w  

  ***********************************************************  

For my part, (as a Master); I will note that the move looks playable, and that it is also the first (move) choice of a few fairly strong chess programs. It is also important to note that this  move does NOT cause an appreciable shift in the computer's overall assessment of the position ... which is probably a good indicator that if Black went astray, it was not on this turn. 

The other side of the coin is that this move DOES represent a permanent weakening of Black's King-side, and therefore might deserve some of the criticism that this particular play has received.  

     [ Another thought was:   >/=  17...Ne5;  with a playable position for Black. ]   

 18.h3 h4;  19.Bh2 Qc7;  20.b4!,  (pawn-lever)   
Trying to break into Black's Queen-side here.   

     [ Or  20.Qc3,  "~"  (about equal). ]   

 20...Bd7;  ('!?')  (developing)    
Maybe the most prudent course for Black here. (Although both 20...b6 and 20...Bf5 might have been slightly better than the course of the actual game, although this point goes unnoticed by our two authors.)   

     [ Or if Black should play:  20...cxb4!?;  then White plays:  21.Qa4,  "~"  ("/\"). ]   

The series of the next few moves should be relatively easy for most players to understand.  
 21.Qd2 Ne5;  22.bxc5 dxc5;  23.d6 Qc8;  24.Qb2 Nc6;    
 25.Nb3 b6;  26.f3!? Re6!?;  {See the diagram - just below.}   
Is Black going for the full point? (Probably.) 

Is this the wisest course that he can take?  (Maybe not.)  {Probably - '?!'}    

  ***********************************************************  

  r1q3k1/p2b1pp1/1pnPrn2/2p5/2P1p2p/1N2PP1P/PQ2B1PB/R4RK1 w  

  ***********************************************************  

But it makes for interesting chess ...  

     [ (>/=)  26...exf3; 27.Bxf3 Qa6;  "~" ]   

 27.Rad1 Qe8;   
Black piles up on the e-file ... (Playing 27...Rb8; was also a thought here.)  

     [ I must also note that Black could play  27...Qa6;  "~"   here - and saved   
       several tempo over the game. ]   

Now White ensures that the e-file will remain closed, thus ensuring that the Rook on e6 has some difficulty finding active play.  
 28.f4! Qc8;   
Perhaps giving notice that Black thinks his Rook on e6 needs a (better) square, and maybe also an admission that his earlier plan (...Re6; and then ...Qe8); wasn't really all that great.  

       [ 28...Nh7!?;  "="  - Fritz 8.0 ]   

 29.f5 Re8;  30.Bg4!? Qa6!?;  {See the diagram - just below.}  
This might be OK, perhaps the strategy of playing defence (with 30...Rd8) did not appeal to the big-time GM here. (Maybe - '?!')   

  ***********************************************************  

  r3r1k1/p2b1pp1/qpnP1n2/2p2P2/2P1p1Bp/1N2P2P/PQ4PB/3R1RK1 w  

  ***********************************************************  

Maybe ... just maybe ... this was the last concrete opportunity for Black to improve his play.  

     [ Possibly this was Black's last (real) chance to play the following   
       continuation of:   (>/=)  30...Nxg4; ('!')  31.hxg4 Qa6;  "~"  {D?}   
       and try to defend his position. 

       (31...Nb4!?; was also worth a look for Black in this position.) ]   

Now White "goes for broke" with an all-out K-side attack.   

 31.Qf2!,   (Maybe - '!!')   
A token of things to come, White disdains the idea of 31.Be2, with simple equality.   

     [ >/=  31.Be2,  "="  according to Fritz 8.0. ]   

 31...Qxc4;  "~"  ("=/+")  (Maybe - '!')   
Now Fritz awards a solid plus to Black, but we all know that the bloody box is too materialistic. (31...Nb4 was also worth serious consideration by Black.)   

 32.Qxh4 Ne5;  ["D" ... or "O?"]   
Heading for the d3-square. 
(>/= 35...Nb4; might have been more accurate, and would not have given White the opportunity to capture the Knight on e5.)  

Now White (re) activates his dark-squared Bishop. 

 33.Bf4! Nd3;  hmmm   {See the diagram - just below.}   
Sokolov goes ahead and sinks the Knight into the d3-square. 
[Now this could be dubious here for Black, Fritz prefers 33...Rac8. (Which makes sense, one of the reasons that Black almost loses this game ... near the very end, a hanging QR on a8 could have led to his downfall.)]   

  ***********************************************************  

  r3r1k1/p2b1pp1/1p1P1n2/2p2P2/2q1pBBQ/1N1nP2P/P5P1/3R1RK1 w  

  ***********************************************************  

Now the position has become utterly scrambled ... its hard to tell who is better here.  

     [ The sensible try of:  (>/=)  33...Bb5!?;   was (also) worth a look here for Black. ]  

White to move in this position.  
 34.Bg5!?,  (Maybe - '!')   
Perfectly logical ... it looks very good indeed.  

(But Raetsky and Chetverik award the '?!' sign.)   

     [ Only later did GM Helgi Olafsson point out the following win, which   
        is very beautiful ... and might be nearly impossible to try and find   
        over the board:   

        >/=  34.Bh6!! Qc3!;  35.Nd4!! gxh6!?;  {This could be best for Black.}   

            (But not: <35...cxd4??;  36.Qg5, "+/-")   

       36.Qxf6! cxd4!?;  37.Qxh6!,  '±'   {Diagram?}   
       when the Black King will probably fall victim to an onslaught from the   
       remainder of White's pieces.   

            ( One sample variation is:  37.Qxh6! Ne5[];  38.Rxd4 f6;   
               If the Black Knight goes to c4, White simply captures it,    
               to be followed by f6, with a mating web around the Black King. 
               39.Bd1!! Rac8!?;  40.Bb3+ Nc4;  41.Qg6+! Kh8;  42.Qxf6+,   
               42...Kg8;  43.Qg6+ Kf8;  44.f6!! Qxe3+;  45.Kh2!, "+/-"  {D?}  
               and Black must start playing "give-away" to prevent mate. )  ]  

Now Black uses his Queen to defend f6 from afar ... 
(34...Nxg4; 35.f6!! is similar to the game, and would just transpose after 35...Qc3; 36.Nd4!) 
 34...Qc3; ('[]' ?)  35.Nd4!, ('!!')   
A problem-like interference. 
(The authors A. Raetsky and M. Chetverik only award this move one exclam, and I follow suit. However, if I were annotating this game for the first time, I would be inclined to give it two, as not even Fritz 9 finds this shot.)   

     [ Or  35.Nd2 Rad8;  36.Be2, "~" ]   

 35...Nxg4;  (forced?)  {simplification}  {See the diagram - just below.}  
Now this looks to be Black's best play ... and Fritz agrees with this assessment of this move.   

  ***********************************************************  

  r3r1k1/p2b1pp1/1p1P4/2p2PB1/3Np1nQ/2qnP2P/P5P1/3R1RK1 w  

  ***********************************************************  

The position has become more and more complex, even the boxes start to lose their way here.  

     [ The move of:  35...Nh7!?;  was another idea for Black.   

        But now (the continuation of): 
        36.Bh5! Ne5;  37.Ne2 Qa3!?;  38.f6,  '±'  (With an attack.)   
        yields White a ton of pressure. ]   

It is White to play here.   
 36.f6!!,  (Maybe - '!!!')    
An extremely brilliant shot. 
(Not even all the programs discover this one, even after 5-10 minutes of machine time.)  

To add to the headache, (for White AND Black!); this is the part of the contest when one {or both} side(s) are generally in time trouble.  

"Black most likely missed this. Now his King is in great trouble." 
  - A. Raetsky and M. Chetverik  

     [ I feel sure that GM I. Sokolov probably expected his opponent   
       to play:  36.Qxg4,  "~"   when Black has no real problems. ]  

 36...Nde5!;   (good defense)   
Black forgets about material, and tries to maintain a watch over the crucial f6 and h6 squares.  

     [ Var. # 36B01)  
       Black should not play:  </=  36...cxd4?!;  37.fxg7 f5;   
       Probably no choice here.   

            ( </= 37...Kxg7?;  38.Bf6+ Nxf6[];  39.Qxf6+ Kh7;  40.Qxf7+ Kh8;  
              41.Qh5+ Kg8;  42.Rf7,  ("+/-")  and mates. )    

       38.hxg4 dxe3;  39.Qh8+ Kf7;  40.g8Q+ Rxg8;  41.Qxc3,  "+/-"    
       and White is winning here.  

      ****************************************************************************  

       Var. # 36B02) 
       Worse {than the above} would be the continuation:   
       </=  36...Nge5?;  37.fxg7 Ng6;  38.Qh6 f5[];   This is forced.  

            ( If  </=  38...cxd4?;  then  39.Bf6!,  "+/-"  {Diagram?}    
              with an unstoppable mate coming up on the h8-square. (R & C) )   

       39.Qxg6 cxd4;  40.Qh6 dxe3!?;   
       Several different programs seem to indicate that this is forced for Black.    

            ( Even worse would be:  </=  40...Re5?;  41.Rxd3! exd3;    
              42.Qh8+ Kf7;  43.g8Q+ Rxg8;  44.Qf6+,  ("+/-")  and mate in 3. )    

       41.Qh8+ Kf7;  42.g8Q+ Rxg8;  43.Qxc3,  ("+/-")   {Diagram?}    
       and Black could call it quits here.    

      ****************************************************************************  

       Var. # 36B03)  
       Simply a nightmare for Black is:  
       </=  36...gxf6?; ('??')  37.Bxf6 Nxf6;  38.Qg5+ Kf8;  39.Qxf6 Kg8;   
       40.Qxf7+ Kh8;  41.Rf6,  "+/-"   Black will be unable to avoid mate.   
       - A. Raetsky and M. Chetverik. (They end their line with Qg5+.)  

      ****************************************************************************  

       Var. # 36B02)   
       Possibly worth investigation was the continuation of:   
       36...Nxf6!?;  37.Bxf6 gxf6;  38.Rxf6 Ne5;  39.Qg5+ Ng6;    
       when White could claim a draw with the simple RxN/g6+,    
       or play for a win with  >/= 40.Rdf1!  ]   

 37.fxg7!,   (en gracio)   
White does not take his eyes off the prize, and continues the attack.   

     [ Possible was: 37.hxg4!? ]   

 37...Ng6; ('!')  38.Qh5 cxd4[];  
This is best or virtually forced, although 38...Re5; was worth some serious investigation.  

     [ But Black certainly cannot allow:  </=  38...N4e5?;  39.Bf6!,  "+/-"    
       with an unstoppable mate.  (On the h8-square.) ]   

 39.hxg4 Re5!;  (nice)   {See the diagram - just below.}  
Black must stop the Bf6 idea (again) by White, as this leads to a mating net.  
(See the note after Black's 38th move, just above.)  

  ***********************************************************  

  r5k1/p2b1pP1/1p1P2n1/4r1BQ/3pp1P1/2q1P3/P5P1/3R1RK1 w  

  ***********************************************************  

Black seems to be holding the {overall} balance ... 

Now White drops another stunner onto the playing field.  
  40.Rf5!! Rxf5;  ('!?'/'?!')   
This should lose ... but so does everything else.   

     [ Var. # 40B01)   
       Or  40...Kxg7!?;  41.Bf6+ Kg8;  42.Rxe5 Qxe3+;  43.Kh1 Qf4;  44.g5, "+/-"   
       and White should win. (An 'eval' of + 2.75 by Fritz 8.);  

    ********************************************************************************************   

        Var. # 40B02)   
        Also of no help was:  (</=)  40...Qc2!?;  41.Bf6!! Qxd1+;  42.Kh2, "+/-"  
        and once more ... White has the unstoppable mate on h8. ]  

The next few blows look to be best ... and nearly forced for both players here.   
 41.gxf5 dxe3;  42.fxg6 Qxg7[];   {See the diagram - just below.}  
This move was probably forced for Black, (just about any other move loses on the spot).  

  ***********************************************************  

  r5k1/p2b1pq1/1p1P2P1/6BQ/4p3/4p3/P5P1/3R2K1 w  

  ***********************************************************  

Despite all the complications, the GM appears to be 'holding the fort,' as we say down south.   

     [ </=  42...fxg6??;  43.Qh8+ Kf7;  44.g8Q+ Rxg8;  45.Qxc3,  "+/-"  

      ****************************************************************************  

        </=  42...Kxg7??;  43.Qh7+ Kf8;  44.Qxf7#. ]     

Now please study the position very carefully. White to move - what play would you make here? 

 43.gxf7+?!,   (Really - '?')   
Amazingly enough, White missed a win in this position.  

[The authors of a book on the English give this two question marks, ('??') but I reserve that appellation for a terrible mistake or even a gross blunder.]   

It IS a shame that White missed all of this, as this would have made this a chess game for all the ages. (Was Gunnarsson short of time?)   

     [ White should have played:  >/=  43.Bf6! fxg6;  {Box?}   
       Practically Black's only move.   

     ***** ***** ***** *****   ***** ***** ***** *****   ***** ***** ***** *****   

            ( Not  </=  43...Qxg6?; 44.Qh8#.  

           ** ** ** ** **  ** ** ** ** **  ** ** ** ** **  ** ** ** ** **    

              Or 43...Qxf6!?; ('?!')  44.Qh7+ Kf8;  45.Rf1 Qxf1+;   
              46.Kxf1 Bb5+;  47.Ke1 fxg6[];  Otherwise Black gets mated. 
              48.Qh8+ Kf7;  49.Qxa8,  "+/-"   with a very easy win. )     

     ***** ***** ***** *****   ***** ***** ***** *****   ***** ***** ***** *****   

       44.Qd5+ Qf7[];  (Otherwise, Black loses the Queen.)   

       45.Qxa8+ Qf8[];    ---> This is (probably) forced. 
       (And not 45...Kh7??; because White replies with 46.Qh8#.)  

       46.Qd5+ Qf7;  47.Qg5,  "+/-"   {Diagram?}   
       and White is a Rook ahead. (A fairly straight-forward win,   
       although White must exercise some caution with Black's two   
       e-pawns still on the board.) ]  

 43...Qxf7;  44.Qe2 Qf2+!?;  {See the diagram - just below.}   
... and the players agreed to a draw in this position.  ("=" maybe "+/=")   

  ***********************************************************  

(The final position.)  

  ***********************************************************  

A game full of interesting tactics, unusual positions, and extremely dense variation trees.   

*************************************************************************************  

   BIBLIOGRAPHY:  

   << "The English with 1...e5." (Reversed Sicilian Lines.) >> 
   By Alex Raetsky and Maxim Chetverik.  (Game # 71, page # 181.) 
   Copyright by the authors (and also the translator), (©) 2003.  
   Published by Everyman Chess.  ISBN: # 01-85744-339-X (flex)   

     [This book on Amazon.com]  [This book on Chess Cafe / USCF booksellers.]  

************************************************************************************* 

   Copyright (c) A.J. Goldsby, 2006. All rights reserved.   

   ½ - ½