Of Attacks

How do we know if an attack is likely to succeed? In other words: On what point should I concentrate the attack? It should be clear to all that it is of no possible use to direct an attack on anything that can move away. Yet beginners frequently infringe this obvious rule.

The diagram below illustrates a mistake frequently made in the choice of an objective, and one which can utterly spoil the whole game, even in its earliest stage. Black has to move, and his game is somewhat hindered by the dominating position of the white queen. The latter prevents the bishop from occupying a desirable square at c5, and also makes the liberating move ... d5 impossible. Therefore it would seem desirable to drive the queen away. But this should only be done if it is not attended by some further disadvantage.

Now the average player is not particularly fastidious in his methods. The queen irritates him, therefore the queen must be repelled one way or the other. He would probably try ... c5.

The result is that the queen selects another good square, for instance at e3 or a4, but Black has not improved matters, for he still can play neither ... Bc5 nor ... d5. On the other hand, irredeemable harm has been done, inasmuch as the black D-pawn now remains "backward." The attack on the queen by ... c5 must consequently be rejected. Sallies such as these, in which short-lived attacks are made by pawns upon pieces, are always of doubtful value. They must unquestionably be avoided if they break up the pawn skeleton, which is formed in the opening, and confine the mobility of the pieces.

Also with regard to maneuvers of pieces, intended solely to drive away an opposing piece, it is obviously essential that the attacking pieces in effecting their purpose should not be made to stray too far afield, lest they become out of play.

I shall delay dealing with the features underlying good forms of attack, both by pieces and pawns, until I have treated of the choice of an objective.

From what we have already expounded, it is clear that the subject of an attack should be incapable of evasion. Should it, in the course of attack, be desired to prevent a piece from being moved, that can only be effected by means of a "pin." A pawn, however, can be held in place either by occupying the square immediately in front of it, or by controlling the latter with more forces than the opponent can bring to bear upon it.

The last diagram exemplifies this. If Black makes the mistake of playing ... c5 as suggested, the backward D-pawn becomes a welcome objective for White's attack. White can keep that pawn back by playing c4 as soon as it threatens to advance, after which he would develop quietly, double his rooks, and bring the queen and dark-squared bishop to bear in a concentrated attack on d6. A position not unlike that in the diagram below will result ultimately, in which Black defends the pawn as many times as it is attacked, but in which White can bring up his E-pawn to the attack, as the D-pawn cannot move away, whilst Black has no further defensive move at his disposal. Play against a backward pawn nearly always develops on these lines, and is even easier when there is no defending bishop of the same color as the pawn.

Such maneuvers, in the course of which pieces are pinned and attacked, are illustrated in the diagram below. The most obvious move, which initiates an attack and at the same time completes the development of the minor pieces, is Bg5. Attacks by means of such devices are so frequent and varied that it will be necessary to treat them at some length, which I now propose to do. I should also add that, with regard to this position, the student will derive lasting benefit from a thorough study of it, and will thus improve his power to judge of the desirability, or otherwise, of obtaining open files, diagonals, doubled pawns, etc.

Matters would be different were Black to allow his king's wing to be broken up without getting rid of White's dangerous pieces by exchanges. Let us consider what happens, if Black takes no measures against Nd5, but only prevents White's ultimate Qf3 by pinning the knight with ... Bg4. White gains a decisive advantage by bringing his queen into play before Black is able to secure himself against the threatened combined attack of queen and bishop, or alternately queen and knight by Kh1, Rg1-g3. I will give two examples of how the whole game now centers on the attack and defense of the points weakened by the disappearance of the G-pawn, and how White pushes home his advantage in the one instance with the help of the bishop, in the other by the co-operation of the knight.

Taking it all in all, we see from the foregoing that the pinning of the black knight can only be injurious to Black if he does not take timely measures to provide against White's Nd5, which threatens to concentrate more forces for the attack on f6 than Black is able to mobilize for its defense.

Beginners, after having experienced frequent trouble through their inadequate defense of this kind of attack, try to avoid their recurrence by making such pinning moves impossible from the first and playing h3 or ... h6 on whichever side the pin is threatened. Apart from the loss of time such pawn moves have various other drawbacks.

With every pawn move it should be considered whether the squares protected by the pawn before it has moved may not need the support of that pawn at a later stage. This is particularly the case with regard to squares in front of the castled king. If one of those pawns pushes on, the squares which have lost its protection frequently offer an opening for a direct attack by the enemy's pieces on the king.

A second consideration is the fact that the advancing pawn itself becomes a target for an assault in which the opponent, moving up a pawn on the next file, brings his rooks into play, or in which he sacrifices a piece for the advanced pawn and the one that protects it, thus robbing the king of the protection he sought to obtain in castling.

The following examples will contribute much to the understanding of this most important subject, the grasp of which will mean a great step forward for the student.

The position below is from the game Scheve - Teichmann, Berlin, 1907.

The following diagram shows a position from the game Marshall - Burn, Ostend, 1907. Strong in the knowledge that the black queen's side pieces are not developed, and can only with difficulty be of assistance in the defense of the king's side because of their limited mobility, White takes advantage of the weakness created by the advance of the black G-pawn to g6, and initiates an immediate assault on the king's stronghold.

In cases where both sides have already castled on the same wing, and the opponent has weakened his position by pushing on one of the pawns of that wing, it is seldom advisable to start an attack with the advance of one of the pawns in front of the king, as the latter's position would be weakened. An attack of this kind is only justified if there is a prospect of concentrating with all speed a superior force before the opponent has time for a counter attack.

The position below illustrates one much favored by "natural" players. Here the advance of the A-pawn would not be a suitable plan of attack for White as his rook is no longer on the A-file, nor could it be brought back in time. In this case White must endeavor to take advantage of the weaknesses at Black's f6 and h6 squares, produced by his move ... g6. This will be the modus operandi: Qd2 followed by Bh6, forcing the exchange of Black's valuable dark-squared bishop. After that the queen in conjunction with one of the knights will attempt to force an entry at f6 or h6, as for instance in the following, the moves of which are taken from a game I once watched and took note of as being most instructive.

A somewhat more difficult case is shown below. Here the advance of the white king's side pawns has undeniably produced weaknesses in the pawn skeleton, and these would be fatal had the black pieces as much mobility as the white ones. But the congestion of Black's pieces on the queen's side makes his defense unwieldy, and White has no difficulty in accumulating his forces on the king's side for the final assault. The prospects are that White will be able to bring home his attack, before Black has a chance of forcing exchanges and of bringing about the end-game, which through the weakness of the White pawns would probably turn to his advantage. The play from the Cohn-Lasker Match, Berlin, 1909, is instructive, and shows how the attack should be conducted in such positions.

In the following position White gets the upper hand because he is one move ahead in the attack on the opposing king's knight. The danger of the concentration of two pieces on this knight lies in the fact that Black is obliged to retake with the G-pawn in case White exchanges on f6 and that thereby the squares f6 and h6 lose their natural protection.

Thus weak points are created of which White may find an opportunity to make use in a manner similar to the ones above. On the other hand, Black gains an open file for his rooks as soon as the G-pawn is out of the way, and the exchange on f6 should therefore not be made until preparations for the occupation of the resulting weak spots are completed. The following variations will throw some light on this rather complex problem.

In the foregoing positions, the important thing to realize is the fact that in a position where both players have castled on the king's side, a dangerous weakness is created if one of the pawns in front of the king is forced to move, and if pressure can be brought to bear upon the two squares which through the move of the pawn have lost their protection.

When the pawns concerned are on the opposite wing to their king, the disadvantages of a premature advance are felt in a different way. The weakness concerns the pawns themselves and not the forces behind them, and is apt to cause the loss of the end-game, particularly of rook end-games. Let us compare the following two positions.

In the one case the formation of black pawns is broken by the absence of the G-pawn, in the other of the B-pawn. The absence of the G-pawn can lead to serious consequences in the middle game, because of the weakness of f6 and h6; it can, however, hardly become awkward in the end-game, as the pawns on the F- and H-files are within the protecting reach of their king.

On the other hand, the absence of the B-pawn is of no consequence for the middle game. There is nothing behind it which could invite an attack. The A-pawn and C-pawn, however, are very weak for the end-game, as they are quite out of reach of the king. I do not wish to imply that Black should have avoided the exchange at his c6 at all cost; such an exchange has always the compensating advantage of opening a file for the rooks, which advantage often means a favorable middle game, as will be readily understood. Further, it is often possible to get rid of the weak A-pawn by pushing it on, and eventually compelling the exchange of the opposing B-pawn, an exchange which can usually be enforced if the rooks have occupied the open B-file. The pawn itself is often useful at c6, in that it can support the advance of ... d5 in the center, should it be desired, or it can, by pushing on, be brought to exercise further pressure on the opposing B-pawn.

The break-up of the pawn position on the queen's side can become awkward in the end-game and sometimes in the middle-game when the pawns can be attacked, and pieces brought to bear on the queen's side without leaving the king's side denuded of forces.

This will be illustrated by the position below, from the game Lazard - Lasker, Paris, 1914. Here the pawn positions on both sides are broken, and the player that occupies the open files first, gets a decisive advantage. In this case it is Black's move. We can conclude at once that White has played the opening badly. He must have lost two moves, for he has still to capture the C-pawn and then, being White, it should be his move. This disadvantage, small as it may seem, with which White has emerged from the opening, is sufficient to bring him into the greatest difficulties. Black, of course, does not defend the pawn by ... Bb7 or ... Bd7, as this would practically reduce the bishop to a tall pawn and, moreover, White, by Rb1 or Rd1, would both attack the bishop and obtain an open file. Instead of that, Black utilities the two moves, which he has, as it were, as a gift in an otherwise equalized position, to bring both rooks on the B-file. This policy allows Black to occupy the seventh or eighth rank at will, and to attack the White pawns from the flank or rear, according to circumstances. This menace hampers the radius of action of the White pieces, as they must always be ready for the defense of the threatened pawns, and this gives Black by far the superior game.

The rooks have no opportunity of making themselves useful until a file has been opened, while the queen often finds an occasion to enter the battlefield on a diagonal. Only in such games can the rooks be made to work at a comparatively early stage of the game, in which the players have not castled on the same sides of the board. For then the advance of the pawns in front of the rooks does not create weakness which endangers their own king. The following middle game from a match of two masters furnishes an example.

We have now seen how the possession of open files reacts on the mobility of the opposing forces, forever increasing their difficulties until the positional advantage is converted into material gain.

Let us now recapitulate the chief points touched upon in the course of our deliberations:

1. Generally speaking, attacks should only be directed to objects which cannot be moved away.
2. If in particular cases the attack is aimed at driving off an opposing piece from an especially favorable post that attack is unwise, if it involves the weakening the pawn position, or if pieces have to take up inferior positions in order to effect their purpose.
3. Pawn moves always create weaknesses, either by leaving other unsupported pawns behind, or by giving opposing pieces access to squares formerly guarded by them, and this more specially so in front of the castled king.
4. Attacks which depend on pawn moves are only justified if overwhelming forces can be accumulated in support, as the advanced pawns might become the object of a counter attack.
5. As pawn moves have very generally some drawbacks, the middle game is the pieces' own hunting ground. As in the opening, the first consideration of sound play in the middle game is to make only such moves as do not reduce the mobility of the pieces.

-- Edward Lasker, 1915, 1918