On a more positive note, 32.d3 c8 33.xe4 (33.d1 xh3 34.xh3 f5+ 35.h2 g4+ 36.h3 f2+ 37.h2 h3#) f5 34.f2 (34.g4 hxg4 35.xd6 d7) xh3+ 36.xh4 g4#.

Of the famous 12...d7 Nimzowitsch says "Black sees in White's kingside pawns a qualitative majority. The text move involves a complicated method of restraint." If 12...e8 then 13.g4 f5 14.d5 e7 15.g5 is stagnant (Nimzowitsch).

He could play 22...Nf3+ 23.Rxf3 (23.Kg2 a6 24.dxc5 dxc5 25.Rfd1 Qh5 26.b7 Bxb7 27.g4 Qh6 28.Nxc5 f5 29.Qg3 Rf6) 22...exf3 24.Qxf3 Bf5 25.bxa7 Bc2 26.Nc1 Be4 27.Qg3 f5 28.Nb3 Rf6 29.dxc5 Rg6 30.Qf2 Rg2+ 31.Qxg2 Bxg2 32.Kxg2 dxc5 33.Ra5 g6 34.Rxc5 Qxa7 35.Rd5 Qa2 36.Nc1 Qxc4 37.Rd4 Qc6+ 38.Kg3 Ra1 39.c4 Qh1 40.Rd8+ Kf7 41.Rd7+ Ke6 42.Rd5 Rxc1 43.Bxc1 Qxc1 44.Re5+ Kd7 45.c5 Qe1+ 46.Kg2 Qe2+ 47.Kg3 Kc6 48.Kh4 Qg2 49.Re8 Kxc5 50.Re5+ Kb4 51.Re8 Kc3 52.Re6 Kd2 53.Rb6 Qg1 54.Rb2+ Kxe3 55.Rb3+ Kxf4 56.Rf3+ Ke5 57.Re3+ Kd5 58.Rd3+ Qd4+ 59.Rxd4+ Kxd4 60.Kg5 Ke4 61.Kxg6 f4 62.h4 f3 63.h5 f2 64.h6 f1=Q 65.Kg7 Ke5 66.h7 Qf6+ 67.Kg8 Ke6 68.h8=N Qh6 69.Nf7 Qg6+ 70.Kh8 Kf6.

Keeping the g-file closed doesn't help. 18.a5 Ng4+ 19.Bxg4 hxg4 20.axb6 gxh3 21.g4 Nh4 22.Qg3 f5 23.g5 Bb7 24.bxa7 Nf3+ 25.Kh1 g6 26.Bc1 Bc8 is a dominant position for Black.

Position at move twelve, aron must have known that it would be best if he concentrates his pieces at the king side. Meanwhile, white's pieces are in disarray!

NO. NEVER.

Okay, maybe that's a bit harsh. However, it's still basically true - in this line (and other variations of the Nimzo that share the same salient qualities, namely White being forced to recapture at c3 with the pawn, and Black refraining from ... d5) it is essentially never to White's advantage to seal the center with e4 and d5, and in fact, it is virtually always to White's disadvantage.

This will be appreciated in light of the pawn formation thus produced. At this point Black (if he has not already) WILL play e5, producing this Pawn skeleton:

click for larger view

Let's consider the strategic objectives on each side in such a pawn structure:

* For Black, there are two salient objectives, one on each wing. The primary is to take advantage of White's c4 pawn, which as we can see in the diagram (unguarded and unable to be guarded by a pawn, and even worse, shielded from c-file protection by the pawn at c3) is very, very sick. To this end, b6, Na5, and Ba6 are to be played. White must then devote two pieces to passive defense of the c4 pawn (while Black's Bishop in particular remains quite mobile, and needs only two moves to shift to King's side action). Moreover, a4 is probably necessary (combined with keeping the Rook stuck at a1 for much of the game) as otherwise, Qd7-a4 threatens, after which White's Queen, Knight, and light Bishop are all glued to c4 (while once again, Black's Queen and Bishop can still get back to the K-side in a jiffy, and will always have the initiative in doing so, since White's guardians can only desert the Pawn AFTER Black's attackers do).

Meanwhile, on the K-side, Black's plan is also clear - f5, opening the f-file by leverage against White's e4 pawn (in fact, it could be argued that ... e4, as in the game, is something of a weakness, since the Pawn there is much looser after f5-f4 than a Pawn at e5 would ever be. Also, White's 11 f4 is probably a mistake, since it makes it that much easier for Black to open a K-side file.) After this, the Rooks are doubled on the f-file and possibly brought to the g- or h-file via f6, before the rest of Black's pieces join in for a strong attack. If White replies to f5 with f3, Black can play f4, and open instead the g-file for an even stronger attack.

* There is precious little that White can do against any of this. His advantages are few - the b-file, the Bishop pair, and a central space advantage. However, none of these is especially exploitable - the Bishop pair is at its poorest in closed positions like this, and in fact, it is debatable if the Bishop pair is even an advantage at all here (mainly based on how much the position revolves around Black's pressure on Pc4, and the fact that White's Bishop is a complete noncombatant in that regard, being on the wrong color squares and not even being able to trade for Black's Knight, since the immured Pc3 blocks this idea hopelessly). The b-file leads to naught, since Black's b6 leaves a White Rook biting on granite. The central space advantage is true, and has some slight defensive value (it at least means Black's Na5 does little offensively but tie up a White piece, since it would take too long for it to assist in K-side action) but there is virtually no advantage to be taken of it with how closed the center remains.

Nor is there really any room for White to obtain any new advantages - in a d5/e4 vs. d6/e5 pawn skeleton, f4 is virtually never good for White, as the f-file is outweighed by the additional weak pawn at e4, and moreover, by the hole at e5 (which in this case is even more significant than Pe4, as Black will place a Knight at e5, making White's illness at c4 even more debilitating - f4 is probably fatal for White here). Nor can White hope to soften up things on the b-file with a4-a5, since Black's Knight blocks the latter square. Should White attempt to drive away this Knight, he can do it only with Nb3 (as Pc3 prevents Bxa5 from ever becoming a reality); to achieve this, Qe2/f1 and Bd3/e2 are first necessary. Meanwhile Black will have played f5, and thus White will need to additionally play f3 (else, once White's Knight abandons c4, Black will win that Pawn by playing fxe4, forcing Bxe4), allowing the aforementioned g-file based attack. Worse, the whole maneuver requires so much time that even if White enforces a5 after all, Black will have time to contest the file successfully anyway.

not necessarily true. if white has control over the queenside, then f4 is a way to open a second front.

a famous example - Karpov vs Unzicker, 1974

Thanks for that. I have the book (in the English translation by Philip Hereford, edited by Fred Reinfeld, with all moves in descriptive notation), so I looked up the game and found that Nimzowitsch says . . . NOTHING WHATEVER about the maneuver!

After 31. Re2 he does say of his next move, 31...Nh4, "Seizes his chance. Black's KP now needs to be defended. If he [i.e., Johner] had limited himself to purely defensive measures, as, say, B-Q2 [i.e., Bd2], a pretty combination would have resulted; namely 31. B-Q2, R-Kt3! 32. B-K1 Kt-Kt5 ch, etc."

What a good way to set up an attack, using the h pawn of the castled king and bringing the queen behind it.

To broadly state 'NO. NEVER.' is general reasoning in situations which require a concrete approach.

<....Nor is there really any room for White to obtain any new advantages - in a d5/e4 vs. d6/e5 pawn skeleton, f4 is virtually never good for White....>

There are exceptions to this generality as well, eg, when White obtains piece play to compensate for the weakness; when he can recapture with pawn, thus keeping Black's pieces from using e5, also keeping the possibility of a breakthrough with e4-e5; when Black is in no position to exploit the weakness of e5, and when tactical considerations enable White to play f4 to his advantage.

<...In short, the strategic idea of sealing the center in this position is, to put it mildly, an extremely faulty one for White, if not outright fatal.>