1 Qh3 h6 2 Rfg4 Kh7
This is all forced.
3 Ne2!
White simply transfers the knight to h5 and carefully avoids the poisonous pawn on e4!
3 ... a5
White also wins after both 3 ... Rh8 4 Nf4 Bd7 5 Rg6 Kg8 6 Nh5 and 3 ... Qf8 (or 3 ... Bd5 4 Nf4) 4 Nf4 followed by Rg6 and Nh5.
JN: Unfortunately 3 ... Rh8 4 Nf4 Bd7 5 Rg6 Kg8 6 Nh5 runs into 6 ... Qxg6, but White could have won earlier by 5 Nh5 Bxf5 6 Rxg7+ Qxg7 7 Qxf5+ Qg6 8 Nxf6+ Kg7 9 Rg4.
4 Rg6 Qf8 5 Nf4 b4 6 Nh5 Rf7
Otherwise White wins even more quickly by Nxf6+.
7 Nxg7 Rxg7 8 Rhxh6+ Kg8 9 Qh5!
The quickest and most precise.
9 ... Rxg6 10 Rh8+ Kf7 11 Rh7+ Qg7 12 Qxg6+ with a quick mate.
This shows how play should have proceeded from the original position, but things would have gone much the same way if, instead of 4 Nxe4? White had played 4 Ne2!, a point of which the reader can easily convince himself. 1 d5?! was only the first step along the wrong path, but 4 Nxe4? was a mistake which could have cost half a point.
So it turned out that Dr Vidmar failed to see something ‘in front of his nose’ in London in 1922, nor did he see it thirty years later when he was writing his book. In this there is without doubt a touch of humour, the humour which Lasker recommended as an addition to every sizeable dose of chess principles.
From attack to endgame
A host of examples could be given to illustrate the transition from an attack on the king into the endgame. It may be a question of heading for a drawn position as a lesser evil to the failure which would result from continuing the attack; it may also be that one variation of the attacker’s combination leads to an ending. However, cases such as these do not really belong here. We are only interested at this point in cases where the attacker deliberately calls off the attack because his success is better assured by a favourable endgame.
The classic exponent of the liquidation of an attack and the transition into the endgame was Pillsbury, so we shall open a series of examples with one of his games.
This position is from Pillsbury-Maróczy, Paris 1900. Although the situation is still far removed from the endgame, White is thinking about it all the same.
1 Ra5
He prepares for the exchange of the dark-squared bishops by means of Bc5, after which there is both an eventual threat of a bishop sacrifice on h7 and the prospect of an ending with White having a good knight against a bad bishop.
1 ... f5 2 Kh1 Rce8
Avoiding the possibility of 3 Bc5 Bxc5 4 Rxc5 Qb6? 5 Rxc8.
3 Bc5 Qc7 4 Bxe7 Rxe7 5 g4
The positional obligation entailed by such a move is balanced by the fact that in this position White has no need to fear the endgame. If his opponent’s dark-squared bishop were still there, the move would be risky.
5 ... Nc6 6 Raa1 fxg4 7 Qh4 g6 8 Qxg4 Rg7 9 h4 Ne7?!
Black defends himself against an attack which is not dangerous and neglects something much more important, namely, the central point d4, which is of decisive significance in many types of ending. Precisely the opposite course was correct, that is, to increase his influence over d4 and at the same time, by 9 ... Qb6. to question White with regard to what his plans were for the dark-square network. In view of the threat of 10 ... Qe3 White would have to play 10 Rae1, after which 10 ... Kh8 11 Re2 d4, opening up the long light-square diagonal, already makes things easier for the defender.
10 Nd4 Nf5?
And now 10 ... Nc6 was correct.
11 Bxf5
A step closer to the ending with knight versus bad bishop. Pillsbury was a specialist in such exchanges.
11 ... gxf5 12 Qh5 Qd8
If 12 ... Be8 13 Qh6 Rg6, White plays 14 Rg1 and is one step further towards an endgame.
13 Rg1 Rff7 14 Qh6 Qe7 15 Nf3
White decides, after all, on an attack instead of simple liquidation by 15 Rxg7+ and 16 Rg1. While attacking. he is always guaranteed a good ending when he wants it.
15 ... Kh8 16 Kh2 Rf8 17 h5
Pillsbury intends to establish his knight on the outpost at g5 and so gives up the line of play leading to an ending which he has followed up to here.
The end of the game, which is no longer germane to our theme, runs as follows: 17 ... Rg4 18 Ng5 Rxf4 19 Nf7+ Qxf7 20 Qxf4 Qxh5+ 21 Kg3 Qe2 22 Kh4 Rc8 23 Rae1 Qxb2 24 Kh3 Rxc3+ 25 Rg3 Rc2 26 Rh1 Rc8 27 Qh6 1-0
If 27 Qh6 Qxe5, then 28 Qxh7+ and 29 Kg2#.
JN: This last section of the game certainly contains some ups and downs, but readers will have to look elsewhere for the details!
In this position from
Bogoljubow-Przepiorka. Pistyan 1922 , the liquidation leading to an ending comes as the direct consequence of White’s combination.
1 Rxh5 gxh5 2 Rxd7 Rxc3
Necessary, since 2 ... Qxd7 allows 3 Qg5+ Kh8 followed by 4 Qf6+ and 5 Bxe5.
3 Nxc3 Qxd7 4 Qg5+ Kh8 5 Qf6+ Kg8 6 Nd5 Re8
If 6 ... h6, then 7 Ne7+? Kh7 8 Nf5? would be incorrect on account of 8 ... Qd1+ 9 Kg2 and 9 ... Rg8+; however, White can play 7 Qxh6 f5 8 Qg6+ Kh8 (or 8 ... Qg7 9 Ne7+ Kh8 10 Qxh5+ Qh7 11 Ng6+, when White wins easily) 9 Nf6 Rxf6 10 Qxf6+ and the ending is won for White.
7 Qg5+ Kh8 8 Nf6 Qd8 9 Qh6 Qd1+ 10 Kg2 Rg8+ 11 Nxg8 Kxg8 12 Qg5+, and White continued with Qxe5 and won the ending.
This position is from the game
Botvinnik-Rabinovich, Leningrad 1934 . At first White’s attack follows a regular pattern.
1 Ng4 Nxg4 2 Qxg4 Bh6 3 Bf6 Rd7
Black could not play 3 ... Rxd3 4 Rxd3 Be4 5 Rg3 Bg6 because of 6 Qh4!, when White remains the exchange ahead. However, 3 ... Rc8 would have been better than the text move.
4 Bf1
4 Bxh7+ would also have been sufficient.
4 ... Qc8
JN: Vuković incorrectly gave 4 ... Qe8 as the move played.
5 Rxd7 Qxd7 6 Rd1 Qc7 7 Bg5!
White breaks off his attack and steers the game towards an ending in which there are two factors which ensure him a clear advantage: control of the d- file and play against Black’s weak c-pawn.
7 ... Bxg5 8 Qxg5 h6
White now continued with the thematic 9 Qd2 allowing the reply 9 ... Be4; the result was that after 10 Qd7 Rc8 11 f3 Bc2 12 Rd2 Bb1 Black’s bishop got in among the small fry on the queenside, and it needed further incorrect play by Rabinovich before he eventually lost. Instead of 9 Qd2, White should have played 9 Qe3, which indirectly maintains his control of the d-file and also prepares for Rd3 and Qd2.
The next example will serve as an excellent conclusion to our discussion of the transition to the endgame. White’s attack has got to the point where it exerts strong pressure on the focal-point h7, and it seems that the end must come quickly. However, appearances are deceptive, for there is no mate and Black can escape into an ending in which he can continue to resist.
Thus, after 1 Bh7+ Kh8 2 Bg6+ Kg8 3 Bxf7+ Qxf7 4 Nxf7 Kxf7, White runs out of checks and is unable to prevent Black from posting his rook on d7, after which it is not at all clear how White can win.
Another possibility is 1 Nh7, and this leads to even greater material gains; Black is obliged to play 1 ... Ne4!, but after 2 Bxe4 dxe4 3 Nf6+ Qxf6 4 Qxf6 e3 the black pawn forces White’s queen to retire, e.g. 5 Qf1 (or 5 Qg5+ Kf8) 5 ... e2 6 Qe1 Kf8, and White can hardly count on victory.
Here we have, as a start, two ways in which the position resolves itself into an ending; the further analysis of the position should soon convince the reader that at every move there is a variety of possible simplifying lines, most of which lead to different types of endings.
The winning process for White is based on the gradual preparation of the pawn structure on the queenside in readiness for the transition into an ending. In this two factors are decisive: the first is that, besides his knight on f6, Black’s queen is also immobilized since it cannot move from e7 because of the need to cover the squares f7 and f6; the second is that Black’s pawns are also condemned to stay where they are. Any move on their part leads
to Black’s losing in the endgame after the manoeuvre Bh7+, Bg6+ and Bxf7+. Accordingly, Black may only move his rook up and down, and even then he is denied the use of the square f8 because of Nh7. Now for the analysis:
1 c4! Rd8
Forced. White threatened 2 cxd5 cxd5 3 Bh7+ Kh8 4 Be4+ Kg8 5 Bxd5 Rf8 6 Qg6+ Kh8 7 Bxf7. If 1 ... dxc4, then 2 Bh7+ Kh8 followed by 3 Bd3+ and 4 Bxc4 wins in the same way.
2 c5!
A step further towards a favourable ending. If 2 cxd5, then Black’s best reply is 2 ... Rxd5!. Also possible here is 2 Bh7+ Kh8 3 Bg6+ Kg8 4 Bxf7+ Qxf7 5 Qh8+ Kxh8 6 Nxf7+ Kg8 7 Nxd8, but after 7 ... dxc4 8 bxc4 b5, Black’s prospects of a draw are more than good.
2 ... Ra8
Now 2 ... Re8 does not work because of 3 Nh7 Ne4 4 Kc1!, when Black is left without the possibility of ... Qd6 (which would have been an effective answer prior to c5) and at the same time there is a threat of 5 Bxe4, etc., now that the white king can hold the enemy e-pawn. If 4 ... f6, then 5 Bxe4 dxe4 6 Qg6+ Kh8 7 Nxf6 and White wins.
3 Ka3 Rb8
It is easy to see that after 3 ... a5 White will win an ending with queen versus rook and knight because the pawn on a5 must fall. If 3 ... b5 White similarly simplifies into an ending, and this time Black loses the pawns on a6 and c6, e.g. 4 Bh7+ Kh8 5 Nxf7+ Qxf7 6 Bg6+ Kg8 7 Bxf7+ Kxf7 8 Qf4 Rc8 9 Kb4 Ke7 10 Qd6+ Kf7 11 Ka5 Kg6 12 Kxa6 followed by Kb7, etc.
4 Ka4 Ra8
If 4 ... b5+, then 5 cxb6 Rxb6 6 Bh7+ followed by capturing the queen at f7 and winning the ending of queen versus rook and knight, much as above.
5 c4! Rb8 6 cxd5 cxd5 7 Ka3!
White threatens 8 Bh7+ followed by 9 Be4+ and 10 Bxd5, which would not have worked at once on account of 10 ... Qd7+. Black’s rook must now go to d8.
7 ... Rd8 8 Bh7+ Kh8 9 Bg6+ Kg8 10 Bxf7+ Qxf7 11 Qh8+ Kxh8 12 Nxf7+ Kg8 13 Nxd8 and White has at last attained a won ending.
The reader may care to investigate this rather studied example further, all the secrets of which lie in the moves made by the pawns on the queenside; the final outcome of the ending rests with them, for without them White is unable to win.
12 The attack on the king as an integral part of the game
We have examined the various aspects of the attack on the castled king and have covered each topic on the basis of present-day knowledge of the art of chess. This material has been presented with all the difficulties and deficiencies of a pioneer, but with the joy of a discoverer who has seen in new knowledge a confirmation of his faith in the values of chess and his own general theory of the game. The attentive reader who has battled his way with the writer from chapter to chapter will have observed the steady widening of the problem of the attack on the king; we began with the basic techniques of combination and have progressed to the kind of understanding which characterizes mastery of the game. The logical path we have taken has justified its call for a high degree of effort in return for new knowledge; its purpose is to show us the significance of the attack on the castled king in the full context of a game of chess. From our examination of the defence against a kingside attack and of the phases of such an attack, it has become clear that a correct attack is conditioned by the situation on all parts of the board and also by all that has happened in the earlier course of the game. An attack on the king is not some strange. exceptional, and daring operation like a grand slam in bridge. It is an action which is both logical and necessary; an action on which a player decides at a certain stage of the game; sometimes, indeed, he must decide on it, if his opponent is not to obtain an advantage or if the position is not to fizzle out into a draw.
Lasker observed long ago that every action is essentially a diversion, a transition from equality to some spatial or temporal commitment. An attack on the king demands a very large diversion both in terms of the number of pieces involved and of the degree of commitment arising out of any pawn moves it entails. A considerable diversion of this kind also logically needs a considerable number of preconditions, as well as a knowledge of how these conditions are interrelated. It follows, therefore, that the attack on the castled king is an extremely difficult operation, which is based on a perfect mastery of all the forms of chess technique. Even in the early stages of the opening the first seeds of the preconditions for an attack on the king are already being sown, and the tense struggle for the central squares is tightly bound up with the successful outcome of the attack or defence. The characteristic features of a position, the weaknesses on both sides, and the elements that affect the endgame – all these accumulate and become intertwined in such a way that at a certain moment they impel the player to decide on the final objective of his play. This is the moment of crisis in the drama of the games of great masters, when it has to be decided whether it is best to continue promoting the preconditions for an attack on the king or to seek a favourable ending. The decision is not easy, for both courses demand not only technique but the ability to weld all the elements involved into a whole; moreover, we are still in the dark about many of the laws governing this process. Chess skill at the present time comprises a mastery of combinative techniques and the possession of mature positional judgement, while the science of the game has cleared up the problem of the centre in the majority of openings. However, we are still a long, long way from a thorough knowledge of the laws governing the methods of integrating all the elements into a strategic whole. And where there is insufficient knowledge one proceeds ‘by feel’, or in other words, one takes risks. The endgame and the attack on the king are two areas about which our knowledge is incomplete, but they are also two directions in which chess can be further explored and through which the ancient game can be raised to a still higher level.
We have said that there still exist areas in the art of chess where our knowledge is incomplete, and many of our readers may ask whether there are not at least some signposts pointing to where new knowledge might be found. The very attentive reader may also recall that mention was made in the Introduction of Alekhine’s and Capablanca’s understanding of the attack on the king and that a discussion on this subject was promised at the end of the book. Indeed, this is the right place for such a discussion, since the most difficult and arcane matters should be dealt with at the end, after everything else has been elucidated.
The discoveries of Alekhine and Capablanca
It is a slightly strange thesis that one should go back to the period before the Second World War and look for the culmination of the art of the attack on the king in the play of Alekhine and Capablanca. For have not Botvinnik and Smyslov, Keres and Bronstein, Reshevsky and Gligorić in their rich creativity in the field of chess discovered all that Alekhine and Capablanca knew and also gone on to add something else? It is indeed true that they have proceeded further in many respects, and perhaps even in their knowledge about the attack on the king they are the equals of the great masters of the past; in their actual games, however, they have not so far put it into practice to the full. If we delve into the whole of chess literature looking for games which serve to illuminate the logical integration of the attack on the castled king and the game in its entirety, then those of Alekhine and Capablanca remain the unsurpassed models; side by side with them may be placed a few of the games of Rubinstein, Keres, and Bronstein, though we shall find scarcely any new discoveries in them; the whole period from Alekhine’s death to the present day has not been particularly fruitful in the sphere of the attack on the king. So one is justified in going back to Alekhine and Capablanca as the first to discover knowledge which to this day has been neither refuted nor superseded.
As far as these actual discoveries by Alekhine and Capablanca are concerned, it can be said that they did not take the form of some new principle but rather depended on a new way of thinking, which proceeded subjectively and therefore in a manner which cannot be expressed or comprehended on a theoretical basis; by these methods they discovered moves which fulfil the chief principle of the attack on the castled king, namely, obtaining the maximum preconditions for an attack with th
e minimum of commitment.
This principle in itself is a simple one and can be easily remembered, but playing in accordance with it is extremely difficult. Here two grades of values must be clearly distinguished: the one based on the value of the individual preconditions and the other on that of the successive commitments undertaken. This must then be resolved into a precisely ordered series of moves, while one must also be ready at any moment to convert the attack into some other thematic course of action – all of which requires considerable mastery.
From a close study of the games of Alekhine and Capablanca one can conclude that they did indeed use a perfected method of thinking in the sense mentioned above; only, in the case of Alekhine this thinking was more analytical and logical, while with Capablanca it was more artistically intuitive. However, the results of this thinking, that is, the actual moves, do not appear to differ greatly to the outside observer. However, Alekhine’s discoveries can be somewhat more precisely defined than Capablanca’s; Alekhine’s output was more abundant, he ‘lived chess’ more deeply, and he was a more passionate analyser of his own games. A particular characteristic of his commentaries is the frequent use of the idea of commitment (just as with Lasker we get the idea of compensation, with Steinitz that of the balance of the position, and with Nimzowitsch that of centralization). One could say that of all masters Alekhine had the most refined sense of the degree of commitment, even though his temperament sometimes led him to be too eager.
Art of Attack in Chess Page 29