Capablanca states that "control of the center is an essential condition for a successful attack against the king."

Control of the center doesn't mean we need dictatorship over all four central squares, as the following game (annotated by newshutz) shows:

[Event "London Chess Classic"]
[Site "London ENG"]
[Date "2011.12.06"]
[Round "4"]
[White "V Anand"]
[Black "Hi Nakamura"]
[Result "0-1"]
[ECO "E97"]
[WhiteElo "2811"]
[BlackElo "2758"]
[Annotator "Fritz 12 (120s)"]
[PlyCount "98"]
[EventDate "2011.12.03"]
[Source "ChessBase"]
[SourceDate "2004.11.03"]
[FEN ""]

1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Nf3 O-O 6. Be2 e5 7.
O-O Nc6 8. d5 Ne7 9. b4 Ne8 10. c5 f5 11. Nd2 Nf6 12. a4 (12. cxd6 cxd6 13. Ba3
Bh6 14. exf5 gxf5 15. Nc4 Ng6 16. b5 Ne8 17. b6 a6 18. Bh5 Rf6 19. Bc1 Bxc1 20.
Rxc1 Nf4 21. Bxe8 Qxe8 22. Ne2 Ng6 23. f4 Bd7 24. fxe5 dxe5 25. Ng3 Qf8 26. Nh5
Rf7 {Postny,E (2591)-Nataf,I (2592) Evora 2006 0-1 (40)}) 12... g5 (12... Kh8
13. f3 Bh6 (13... g5 14. Nc4 g4 15. fxg4 dxc5 16. bxc5 fxe4 17. d6 cxd6 18. g5
Ne8 19. cxd6 Ng6 20. Nxe4 Be6 21. Ba3 Rf4 22. Rxf4 exf4 23. Rb1 b6 24. Bb2 Rc8
25. Bxg7+ Nxg7 26. Qd4 Rc5 27. Nxc5 {1-0 (27) Petran,P (2386)-Werner,D (2362)
Budapest 2004}) 14. Nc4 Bxc1 15. Qxc1 f4 16. a5 g5 17. b5 Ng6 18. b6 dxc5 19.
Nb5 Ne8 20. Nxa7 Bd7 21. Qb2 cxb6 22. axb6 Qf6 23. Nb5 Rb8 24. Nc7 h5 25. h3
Nd6 26. Rfc1 Nxc4 27. Rxc4 {Schussler,H (2420)-Ivarsson,J (2355) Stockholm
1980 1-0 (37)}) (12... f4 13. Nc4 Kh8 14. Ba3 Ne8 15. a5 Ng8 16. f3 Rf7 17. Qb3
Ngf6 18. b5 Nh5 19. Rfc1 dxc5 20. Bxc5 Nd6 21. b6 Nxc4 22. Bxc4 cxb6 23. axb6
a6 24. d6 Rf8 25. Nd5 Bd7 26. Nc7 Rb8 27. Ne6 {Lekic,D (2390)-Filipovic,M 
(2121) Budva 2009 1-0 (34)}) (12... fxe4 13. Ndxe4 Nf5 14. Bg5 h6 15. Bxf6 Bxf6
16. Bg4 Bg7 17. b5 Rf7 18. c6 bxc6 19. bxc6 h5 20. Bh3 Ba6 21. Re1 Rb8 22. Rb1
Rxb1 23. Qxb1 Bh6 24. Nb5 Kg7 25. Ng3 Nxg3 26. hxg3 Bd2 27. Re4 {Gleizerov,E 
(2405)-Temirbaev,S (2375) Uzhgorod 1988 1/2-1/2}) 13. Nc4 $146 ({fritz opening
book variations} 13. a5 f4 14. g4 (14. f3 Ng6 15. a6 b6 16. Bb5 g4 17. Bc6 Rb8
18. Nb5 g3 19. h3 Bxh3 20. gxh3 Qc8 21. Re1 Qxh3) 14... Ng6 15. f3 h5 16. h3
Rf7 17. Rf2 (17. Ba3 Bf8 18. Nc4 Rh7 19. Kg2 Be7 20. b5 Qf8) 17... Bf8 18. Ba3
Rh7 19. Nf1 Be7 20. Rc1) 13... h6 $6 {wastes a tempo as ...f4, f3 is going to
happen anyway.  IMHO,13...f4 was called for. OTOH,  Nakamura is a much better
player than I. The engines give 13...h6 as black's best move, but white has a
clear advantage} (13... f4 {gets back into a main line} 14. f3 Ng6 15. Ba3 Rf7
16. b5) (13... h6 14. f3 fxe4 15. fxe4 Ng6 16. Ba3 Nf4
17. Bf3 b6 18. cxd6 cxd6 19. Rc1 g4 20. Be2 h5 21. b5 Ne8 22. g3 Nh3+ 23. Kg2
Rxf1 24. Qxf1 Bh6 {0.58/25}) 14. f3 f4 15. Ba3 Ng6 16. b5 dxc5 17. Bxc5 Rf7 18.
a5 h5 19. b6 g4 20. Nb5 cxb6 21. axb6 g3 {Though the pawn chain pointed at
White's king looks threatening,  Nakamura is a long way from finishing his
kingside attack. Anand has a tremendous advantage, but needs to play actively
to bring the point home.} 22. Kh1 $2 {Not a losing move, but much too timid.
Anand was trying to secure his position by making sure that ...gxh2 did not
come with check. Preventing ...gxh3 with 22.h3 would allow the thematic ...
Bxh3 sacrifice that is usually bad for white.} (22. Qc2 {Does not look all
that active, but keeps all of White's threats in the air, and positions the
queen for invasion of Black's position along the c-file, frees the Rf1,
provides defensive support to g2, support for the Bc5, and support for a
knight posted on f5 (via d6, which is not a square the knight wants to rest on,
as it blocks the d-pawn, the Bc5, and the other knight.} a6 (22... gxh2+ 23.
Kh1 {is safe enough, as both black knights are far from g3. White might be
able to take on h2, but that requires a lot of calculation, that might be
better spent on other lines.}) 23. Ncd6 Rd7 24. Nf5) 22... Bf8 23. d6 {I think
this is premature. White can maintain control of d6, so the pawn is not going
to be blockaded, soon.} ({Fritz likes} 23. Rxa7 Rb8 24. Ncd6) 23... a6 24. Nc7
Rb8 25. Na5 Kh8 26. Bc4 Rg7 27. Ne6 Bxe6 28. Bxe6 gxh2 29. Nc4 $2 {Where is
this knight going? If Anand felt the need to shore up the d6 pawn, then I
think 29.Ra2 would have been better.} ({Fritz 12:} 29. Bh3 Nh7 30. Qd5 Nf6 31.
Qd3 Nh7 32. Nc4 Ng5 33. Bf5 Nf7 34. Qd5 Qf6 35. Qe6 Qg5 36. Ra2 {2.39/25})
29... Qe8 30. Bd5 $2 (30. Bh3 {would stifle any black attack on the kingside,
and keep control of c8}) 30... h4 31. Rf2 h3 32. gxh3 Rc8 33. Ra5 Nh4 34. Kxh2
Nd7 35. Bb4 Rg3 36. Qf1 Qh5 37. Ra3 a5 38. Be1 Rxc4 $1 {This exchange
sacrifice allows Nakamura to activate his dark square bishop and bring it into
the attack.} 39. Bxc4 Bxd6 40. Rxa5 Bc5 41. Be2 $2 (41. Rxc5 {returning the
exchange this way allows Anand to keep the Rf2} Nxc5 42. Be2 Ne6 43. Kh1 Ng2
44. Qxg2 Rxg2 45. Rxg2 {and things are pretty grim for white, but if he can
exchange off the rest of the pawns, he might be able to hold.}) 41... Bxb6 42.
Rb5 $2 Bd4 (42... Bxf2 $1 {I was screaming at my computer as I watched this
game live. It also would have worked on the previous move, but I did not see
it then.} 43. Bxf2 Nxf3+ 44. Bxf3 Qxf3) 43. Bd1 $2 (43. Rxb7 {would have
lessened the damage.}) 43... Bxf2 {finally, this move would have worked on move 41 and 42} 44. Bxf2 Nxf3+ 45. Bxf3 Qxf3 46.
Rb1 Rg6 47. Rxb7 Nf6 48. Rb8+ Kh7 49. Rb7+ Kh6 {White must give up material,
or the e4-pawn will fall and white cannot stop the black pawns.} (49... Kh6 50.
Rb6 (50. Rb8 Nxe4 51. Rb2 Rg3 52. Re2 Qd3 53. Bxg3 Nxg3 54. Qf2 Qxe2 55. Qxe2
Nxe2 56. Kg2 Nd4) 50... Ng4+ 51. hxg4 Rxb6) 0-1

In the above game, it is evident from move 20 onwards that white attains control of the central d5 square and more than half of the d-file. However, it is black's prevention of white's control over the e5 square (by blockading f2-f4 and keeping a pawn at e5) that grants black the necessary space to launch a kingside attack. In this sense white has no counterplay on the kingside. (Black also controls d4 with his e5 pawn).

What if black had no control of the center? It is widely accepted here that even a violent attack against the king is favored over control of the center. How violent does black's attack have to be when white unambiguously controls at least three of the four central squares?


From the accepted answer you linked to:

The center is the crossroads of the board. Controlling it will give you access to every other part of the board. At the same time, it will drive a wedge in the opponent's position that hampers communication between king and queen side.

This is a general rule which is widely applicable. However, in this case, the e- and f-pawns block White's access to the kingside. And communication between king- and queenside isn't really necessary for Black: all his action takes place on the kingside.

So basically this is just the exception that proves the rule.


I will tell you three rules that say when space isn't useful:

  1. All the general rules work until there's something concrete. That means it is useless to have complete dominance over the center if you're losing because of a concrete line. Then the question isn't "How violent is the attack?" but "Is there a win?".

  2. Center only creates space for your pieces that has to be used. If the space isn't used properly (your pieces have to be well-placed), center won't help you. It actually creates only more weaknesses.

  3. Your pieces should be prepared for a fight when your opponent starts the attack on the king. Then you should act as quickly as possible. If you find yourself still developing or improving your pieces while your opponent's attack is already up and coming, it's too late. I think the game you mentioned is a good example of it. Anand was too slow in the center. He had a good-looking pawn on d6, good-looking center and an open c-file, but that was all he had. In the meantime Nakamura was already developing his play on the white's king.

It's hard to answer directly the question "How violent should be the attack?", because firstly in chess every position bears different amount of violence and secondly there's no unit for violence... :-)

I intentionally don't mention a weak center that could be destroyed in the cases above, because that's off topic here (we're talking about an attack on the king).



I think you have to consider the time and place Capablanca made his comments - 1921. Although Capablanca was a brilliant player, Chess Strategy has come a long way in 96 years, and this exact position is dealt with on page 104 of John Watson's "Secrets of Modern Chess Strategy", in the chapter titled "Rule-Independence", The Demise of the General Rule.

He goes on to discuss the modern idea by White of 9.Ne1 Nd7 10.f3 f5 11.g4!? and white attempts to stonewall black's dogmatic kingside attack with h4 and g5, and have a free hand on the queenside with a c5 break.

Just something to consider when you are in post mortem and cannot fathom how you got rolled by a plan that seemingly violated every dogma you have learned about the centre. I do cringe when I see "widely accepted" used to bolster some dogmatic saying from one of the old masters. In today's game, we look for ways to prove it wrong.

Good Luck and Good Chess!

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.