Art of Attack in Chess

by Vladimir Vukovic

  JN: Actually this just wins for White.

  2) 20 ... Rg4 (covering the square f4, lessening the danger on f7, and preparing ... h5-h4) 21 a6! (logically switching to the positional motif on the queenside and thus exploiting the slower tempo of Black’s attack betrayed by the move ... Rg4) 21 ... b6 (after 21 ... 0-0-0? 22 axb7+ Kb8 23 dxc6 Black would be in great difficulties) 22 dxc6 h5 23 Kh1! (having made good his superiority on the queen’s wing, White goes over to direct defence; he returns the extra piece in order to simplify into a favourable ending) 23 ... h4 (Black would recover less material by 23 ... Nxe2 24 Ncxe2, since then 24 ... h4 is met by 25 Qf1, when White clarifies the position at the further expense of his e-pawn) 24 Bxf4 exf4 (the alternative 24 ... Rxf4 25 Qg2 is equally in White’s favour) 25 Qf1 Rxg3 26 Qxh3 Rxh3 27 Kg2 f3+ 28 Kxh3 fxe2 29 Re1 with a clear advantage to White.

  JN: 23 ... Nxe2 24 Ncxe2 h4 25 Qf1 is still not totally clear after 25 ... Rxg3 26 Nxg3 Qg4, but White can improve earlier by 25 Ng1! Rxg3 26 Qxg3.

  Now we return to the game after 18 Qd2.

  18 ... N6h5?

  Commentators have praised this move, which, however, could have allowed White to escape. Correct was 18 ... h5! which excludes the possibility of simplification and wins back the piece without giving up the attack.

  JN: I do not see how this wins back the piece. White can play (amongst other promising ideas) 19 Bc4 h4 20 Bf1 Qd7 21 dxc6 bxc6 22 Nf5, and Black has nothing to show for his sacrifice.

  19 Qf2?

  White misses the chance to liquidate the position which Maróczy pointed out: 19 Nxh5! Qxf3 20 Bxf4 (forced) 20 ... gxf4+ 21 Ng3 fxg3 22 h3! and Black’s attack loses its power. He must be satisfied with winning the exchange by 22 ... Qf2+ 23 Qxf2 gxf2+ 24 Kxf2 Bh4+ 25 Ke2 Bxe1 26 Rxe1 with a probable draw.

  19 ... g4 20 Kh1

  Interesting is 20 Nxh5 Nxh5 21 Qg2 (or 21 f4 g3!) 21 ... gxf3! 22 Qxg8+ Kd7 when White’s queen has not a single square on the g-file from which to defend g2.

  20 ... gxf3 21 Rg1 Bh4 22 Bxf4 exf4 23 Nxh5 Bxf2 24 Rxg8+ Ke7 25 Nxf4 Qh4 26 Rg7

  Or 26 Rxa8 Bg3 27 h3 Bxf4.

  26 ... Kf8 27 Rxf7+ Kxf7 28 dxc6+ Kg7 29 Ne6+ Kh8 30 Nd5 bxc6 0-1

  This game won the Second Brilliancy Prize.

  In this position Black is to move and must discover a defence against White’s two threats of Nh5 and Nf5. It is fairly easy to reach the conclusion that there is not a large choice; only two moves are worth considering – 1 ... f5 and 1 ... Kh8. The first appears more natural than the second, but in fact only the second provides an adequate defence. If 1 ... f5?, White replies 2 Bxf5!, after which 2 ... Rxf5 does not work because of 3 Nxf5 gxf5 4 Qg5+ Kf7 5 Qf6+ Ke8 (or 5 ... Kg8 6 Rf3) 6 e6 and loses; nor is 2 ... gxf5 satisfactory, e.g. 3 Nh5 Rd7 4 e6, when the rook is lost, since if 4 ... Re7, then 5 Rf3 is decisive; the one course remaining, therefore, is to decline the sacrifice by 2 ... Ne7, but then White plays 3 Bd3, threatening 4 f5 as well as 4 Nh5.

  1 ... f5? is weak because it ‘opens up the position’, with the result that White’s queen becomes even more effective and is always creating new threats. On the other hand, 1 ... Kh8, which is also primarily a form of direct defence, should enable Black, provided he can survive the immediate pressure, to make good his material advantage and territorial superiority on the queenside. Let us now examine this second move.

  1 ... Kh8! 2 Nf5

  After 2 Nh5 Rg8 3 Nf6 (otherwise 3 ... Bf8!) 3 ... Rg7 4 Rf3 Black can just defend himself in time by 4 ... Qc3 5 Rd1 Be7 and if 6 Rh3, then 6 ... Bxf6.

  2 ... Rg8 3 Rf3 gxf5

  Forced, since even 3 ... Bf8 does not prevent White from sacrificing his queen on 4 Qxh7+ and mating with the rook on the h-file.

  4 Rh3 Rg7 5 Bxf5

  5 ... Qd2!

  Now White threatens 6 Bxh7 and 7 Bg6+ followed by mate with the queen at h8, but Black is a knight up at this stage, so the possibility of defence by giving back material arises. As a result of this increase in his resources, Black’s defence turns suddenly from direct to indirect, only in the end to return again, after the ‘detour’, to direct defence.

  6 g3

  If 6 Rg3, then 6 ... Rdg8; anything else is answered by the same manoeuvre as now follows.

  6 ... Bxe3 7 Bxh7 Bxf4! 8 Qh5

  Or 8 Qxf4 Qxf4 9 Bf5+ Kg8 10 gxf4 Nd4 11 Bd3 c5, when Black stands better.

  8 ... Bh6 9 Bf5 Nxe5 10 Qxh6+ Qxh6 11 Rxh6+ Kg8 and Black has a winning advantage.

  JN: Although Vuković is correct with his general principles. the above example is hardly a convincing demonstration; whatever Black plays, he cannot hope to defend the position with most of his pieces stuck far away on the queenside. After 1 ... Kh8 White can play 2 f5! Nxe5 3 Nh5 Rg8 4 fxg6 fxg6 5 Qg5! Nd7 6 Rf7 with a decisive attack, for example 6 ... Qc3 7 Rd1 Rdf8 8 Rxd7 Qxd3 9 Qe5+ and mates. Moreover, even one of Vuković’s own lines favours White: after 2 Nh5 Rg8 3 Nf6 Rg7 4 Rf3 Qc3 5 Rd1 Be7 White need not allow the capture on f6 but can simply play 6 Qh4! threatening 7 Rh3. Black has nothing better than 6 ... h5 7 Qg5 Bxf6 8 exf6 Rh7 9 f5 Qe5, but after 10 h4! he is clearly in serious trouble as his rook on h7 is totally out of play.

  In this example too we observe how a successful defence is flexible and unites a diversity of elements. Direct and indirect defence alternate, the attacker’s sacrifice provokes a counter-sacrifice from the defender, and in the end the basic soundness of Black’s position proves itself.

  This position is from the game Rossolimo-Pachman, Hilversum 1947. It is Black’s move; he has a promising game and ought really to continue along positional lines, e.g. 1 ... a5 2 a4 e4 3 Qe2 b6, etc. Instead, he decides to ‘invest his advantage in an attack on the king’, though the conditions for this are not particularly suitable.

  1 ... f4!? 2 exf4 exf4 3 c5!

  Black opened his attack at the cost of his advantage in the centre. Now White, by ridding himself of his doubled pawns, eliminates the last element of Black’s positional superiority, with the result that the latter has only his kingside attack left.

  3 ... Qf7 4 Be4

  As a general rule, the fianchettoed bishop protecting the castled position should not be touched; but here it is a question of gaining a tempo in order to play cxd6 before Nd2. If at once 4 cxd6, Black replies 4 ... f3, attacking the bishop on g2.

  4 ... Qh5

  If 4 ... f3, White is not obliged to go in for the double-edged 5 Bxh7+. He also has 5 cxd6 cxd6 6 Nd2 Bg4 7 Qc4, forcing the exchange of queens.

  5 cxd6 cxd6

  Black’s position is not strong enough for the uncompromising 5 ... f3 6 Nd2 Bg4. White would continue 7 Rab1, not allowing his opponent to escape from his material commitments; at the same time, he has a direct defence at hand in the shape of an attack on Black’s queen by Rb5. After, for instance, 7 ... Rf6 8 Rxb7 Black’s obligations are too great in proportion to the force of his attack. Consequently, if 8 ... Rxd6, White can permit himself 9 h4, a weakening which does not entail danger, since 9 ... g5 can be countered by 10 Rb5. If instead 8 ... Qh3, then 9 Nxf3 Bxf3 (otherwise 10 Nh4 and Bg2) 10 Bxf3 Rh6 11 Re1 leads to a win for White.

  6 Nd2 Rf6

  Now 6 ... f3 is no longer possible.

  7 Bf3 Bg4

  8 Bxg4?

  This exchange involves the loss of a tempo. White should have switched at once to a counterattack on the queenside, where he is the better placed, for example 8 Rab1! Rh6 9 h4. Then, as we saw above, 9 ... g5 does not work, nor does 9 ... a6 because of 10 Rxb7 g5 11 Qc4+. Lastly, there is 9 ... d5 10 Rxb7, and again 10 ... g5 is insufficient since after 11 Bxd5+ Kh8 12 Bf7 Be2 13 Qf5 White wins.

  This is an example of how indirect defence, by clearing the ground on the queenside, makes it easier to revert to direct defence at the critical moment.

  8 ... Qxg4 9 Qf3 Qh3 10 Ne4!?

  Here too 10 Rab1 would probably have been better.

  10 ... Rh6 11 Qg2 Qh5 12 g4

  After a period of timid play White aims to save himself by blocking the position.

  12 ... Qh4 13 Rab1 Rg6?!

  This mo
ve and, in particular, the next, are not the best. Black should have played 13 ... Re8 14 Rfe1 (otherwise ... Rxe4) 14 ... Re7, which would have enabled him to retain a slight advantage.

  14 h3 h5?

  Admittedly, this forces f3, which shuts the long diagonal to White’s queen, but on the other hand, it undertakes an attacking obligation for which there is insufficient justification. The point is that Black is obliged to obtain some concrete advantage before his pawn at h5 becomes an anxiety and he is forced to strengthen White’s position by exchanging it on g4. After this mistake White is the better placed and should in the end make his presence felt on the queenside.

  15 f3 Na5

  16 Kh2

  Though this does not mar White’s position much, it would have been stronger to have played 16 Rb5! Nc4 17 Qf2!, when Black’s discomfort over the h-pawn is revealed in various ways. Thus, 17 ... Qxf2+ 18 Rxf2 hxg4 19 fxg4 leaves two pawns en prise, one of which will obviously fall, while if 17 ... Qxh3, then 18 Ng5 Qg3+ (or 18 ... Rxg5 19 Rxg5 hxg4 20 Qh2 Qxh2+ 21 Kxh2 Ne3 22 Rb1 gxf3 23 Kg1 is good for White) 19 Qxg3 fxg3 20 gxh5 a6 21 hxg6 axb5 22 Re1 and White is the better placed.

  This line, wherein the union of direct and indirect defence is especially close, would have been a much better way of handling the position than the passive one actually chosen.

  JN: In the 17 ... Qxh3 18 Ng5 Rxg5 variation, 21 ... Nd2! leads to a rook and pawn ending which is winning for Black. White can improve earlier in this variation, but only to the extent of holding the balance. It is therefore doubtful if 16 Rb5 was any stronger than the move played.

  16 ... Nc4 17 Rg1?

  The decisive error. He should have played 17 Qf2; then 17 ... Qe7 (17 ... Qxf2+ eventually loses a pawn) 18 gxh5 Rh6 19 Rg1 Rxh5 20 Rxb7! Rxh3+ (or 20 ... Qxb7 21 Nf6+) 21 Kxh3 Qxb7 22 Qh4 leaves the advantage with White.

  17 ... d5! 18 Qe2

  This loses material, but even the more stubborn 18 Nc5 would not help much.

  JN: Actually White can resign after the reply 18 ... Ne3.

  The deciding point here is that 18 Qf2 does not work because of 18 ... Qxf2+ 19 Nxf2 Nd2!, and White loses the exchange.

  18 ... hxg4 19 fxg4 Re6 20 Qf2 Qe7 21 Rxb7 Qxb7 22 Nc5 Qe7 23 Nxe6 Qxe6 24 Qxf4 Rf8 25 Qg3 Nd2 0-1

  In this position White played:

  1 Re3

  He is clearly preparing to deploy the rook on g3 or h3 in order to attack the opposing king. Moves of this type belong on the whole to a rather more advanced stage of the game, when all the other pieces have been developed; they are also good in closed positions, where the opponent has no chance of counterattacking on an open file. However, even here this sortie on the part of the rook does not entail a very serious commitment for White, since the rook can turn back if it does not succeed in its mission.

  1 ... Bc6 2 Rg3 Kh8

  This parries the threat of 3 Bh6 without weakening the castled position. 2 ... Be4 3 Bh6 Bg6 is not so good because of 4 Bxg6 hxg6 5 Be3, when White has attacking chances along the h-file.

  3 Be3 Be4

  Black is still applying prophylactic direct defence, since the conditions are not yet favourable for a counterattack; the exchange of light-squared bishops ought to lessen the force of the attack and at the same time clear the squares on the c-file for an eventual counterattack.

  4 Rh3?!

  This threatens 5 Bxe4 followed by Qh5, but it does not take sufficient account of temporal factors, with the result that the initiative now passes to Black. It would have been better to contest the c-file by 4 Rc1.

  4 ... Qd5 5 f3

  The second logical step along a dangerous path: if the attack does not succeed, White will obviously suffer from the lack of communication between his rooks.

  5 ... Bxd3 6 Qxd3

  6 ... Bd6?

  Up to here Black has correctly carried through a direct defence without weakening his pawn position, at the same time keeping his trumps in reserve (the c-file and control of the central post at d5). All this has been entirely appropriate to the fairly modest commitments taken on by White. Now he mistakenly supposes that the moment for a counterattack has arrived and makes a careless move, which White could have punished. The situation required direct defence to be continued for one more move, namely, by a renewed consolidation of the king’s position by 6 ... Kg8. The reason is that Black may eventually be forced (by Bg5) to play ... g6, after which Bh6 would threaten the rook protecting f7. This point can be covered by the king after ... Kg8; consequently, Black may then use both his rooks on the c-file, while retaining a sufficient direct defence of the castled position by ... g6, if necessary. 6 ... h6 would be weaker on account of 7 g4, while 6 ... Rc7 is also incorrect, e.g. 7 Bg5 h6 8 Bf4 followed by g4.

  7 Bf4?

  This throws away the chance of victory which was afforded by 7 Nd7!. Owing to the threat of 8 Nxf6, Black loses the exchange. He is not helped by 7 ... Bxh2+ 8 Rxh2! (not 8 Kxh2? Qxd7 9 Bg5, since Black defends himself by 9 ... Qc7+ and 10 ... Qc2!) 8 ... Qxd7 9 Bg5, when he either loses his knight or is mated. After the erroneous text move Black reasserts his advantage.

  JN: White has an even simpler win by 7 Bg5!, after which Black must surrender a piece to avoid being mated on h7.

  7 ... g6 8 g4

  Black was threatening 8 ... Nh5; if 8 Bh6, Black can of course reply 8 ... Bxe5.

  8 ... Rc7 9 Qe3 Rfc8

  Black’s counterattack is growing rapidly in strength.

  10 Bg5 Bxe5 11 dxe5 Ng8

  This is not a defensive move but a preparation for later action against the bishop by means of ... h6.

  12 Rg3

  White concedes that his attack has been beaten off and tries to provide some resistance on the second rank. After 12 Bf6+ Nxf6 13 exf6 Qc5 Black has the better ending.

  12 ... Rc2 13 Rg2 Rxg2+ 14 Kxg2 Rc2+ 15 Kg3 Kg7! 16 Bf4

  The bishop is in danger! A sad fate for an unsuccessful attacker!

  16 ... h6 17 h4 h5! 18 gxh5

  Otherwise Black opens up White’s king position by ... hxg4.

  18 ... Ne7 19 Qb3

  19 h6+ Kh7 would not help White either.

  19 ... Nf5+ 20 Kg4

  He could have put up a better fight by 20 Kh3, though even then Black’s game is overwhelming.

  20 ... Rc4 21 Re1

  If 21 Rd1, Black can play 21 ... Qxd1!.

  21 ... Rxf4+ 22 Kxf4 Qd4+ 23 Re4 Qf2! 24 Kg5 Qd2+ 25 Rf4 Qd8+ 26 Kg4 Qxh4#

  If one leaves aside Black’s unpunished mistake on the sixth move, then the whole course of the game is a good illustration of the relationship between the scale of the attacker’s commitment and the method of defence employed. In the first phase White is not heavily committed by his rook’s excursion and Black carries out a cautious direct defence without compromising his king position. He also maintains his hold on the c-file and the centre and eliminates one enemy unit – the light-squared bishop. With f3 and g4 White has already taken on greater obligations and so the time is ripe for indirect defence by means of an action along the c-file. As soon as another unit is eliminated (the knight on e5) it is all over with White’s attack, whereas the counterattack grows from move to move, with the lack of communication between the white rooks as the main cause of White’s misfortune.

  Defence by a central thrust

  Attacks on the castled king in which the situation in the centre does not play a greater or lesser part are extremely rare, and as a rule it is precisely the soundness of the attacker’s central formation which provides the basis for his attack. If this formation is not sound enough, the defender may use indirect defence against it in the first instance, aimed against the central formation’s most vulnerable point.

  Why is it that the centre is quite so important and sensitive in the particular case of an attack on the castled king? Because this kind of attack is one of the most complex and many-sided operations in chess, demanding the maximum collaboration of the pieces and a control over all possible consequences right up to the endgame. The centre is the most versatile area on the ch
essboard; it provides the best conditions for the full co-operation of the pieces – the basis on which positional play is founded and a prerequisite for a favourable endgame. Many-sidedness is the common property belonging to a sound attack on the castled king and to the centre, and this accounts for their mutual interdependence. Correct defence must also be many-sided – this we have seen in the last series of examples – and so each ‘little crack in the centre’ is more important to the defence than a variety of other weaknesses on the attacker’s side.

  In some of the examples so far we have seen the centre acting as the basis of an attack, while in others the defender has employed indirect defence based on destroying the attacker’s centre by sacrificing a piece. We have noticed particularly the case where the defender strives to centralize his pieces – giving a position which as a rule makes possible a combination of direct defence and the indirect form on the other flank. It still remains for us to look at the case of defence by means of a pawn thrust against the attacker’s central formation.

  Before we pass on to examples, we must be clear that we are concerned here with a central thrust of a particular kind. For why should such a thrust be significant just at the critical phase of the attack? Why did the player not employ it earlier? The point is precisely that we are dealing here with cases where in fact the central thrust is not useful before the crisis, because the attacker has good replies to it. Only when he has increased his commitments and deployed his pieces for the attack does the moment for a thrust in the centre arise; and very often this occurs at a precise, given moment, not a move earlier or later.

  The following famous game provides a good example of a defensive thrust in the centre.

  Alekhine - Botvinnik

  Nottingham, 1936

  Sicilian Defence, Dragon Variation

  1 e4 c5 2 Nf3 d6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 g6 6 Be2 Bg7 7 Be3 Nc6 8 Nb3 Be6 9 f4 0-0 10 g4