Stalemate: the long and the Short of it

Pros and cons of the stalemate rule 

I’ve been a fan of the irrepressible English GM Nigel Short for a long time, so when Nigel expresses a forceful opinion about chess I take it seriously. Those of us who recently enjoyed his online commentary on the Norway Chess 2014 tournament were treated (again) to Nigel’s strongly held opinion about what he might politely call a chess anomaly, or more colloquially “a stupid rule”, namely the rule that scores stalemate as a draw.

Nigel holds forth for the idea that a stalemate should be scored as a loss for the stalemated player. Since he has promoted many good ideas – such as critiquing the “football” scoring of three points for a win by pointing out that the victory then goes to the player with the most losses (just as the tie-break rule favoring “most wins with black” similarly favors the player with most losses as White) – I thought his “stalemate = win” campaign was worth some thought. I’d therefore like to follow up Nigel’s remarks by inviting to a colloquy about the pros and cons of the stalemate rule.

In practice, the stalemate idea is often used to rescue the weaker side from a probable loss. The weaker side typically tries – often by giving away material – to force a position where the stronger side has no choice but to stalemate the weaker. Usually this involves an oversight of the drawing line (a blunder, really) on the part of the stronger side, who thereby contributes to the stalemate, if only through carelessness. (Nigel feels this should be rewarded with a full point.)

Nigel’s problem with the idea of stalemate – if I understand him right – is that it unfairly overturns the good work that a player has put into obtaining a decisive position, and rewards a losing player’s inferior play by allowing him to resort to a cheap trick to save his position. I agree that a reasonable argument can be made against the stalemate rule: after all, when our opponent is stalemated, he is in fact incapacitated.

It seems only logical that when we’ve incapacitated the opponent the game shouldn’t be scored as equal. We can also argue that, since each side is required by the rules to make moves in turn (passing on your turn is not permitted), when a player is unable to make a move as required by the rules, he shouldn’t be rewarded for it with a half point. Nigel brought out the fact that being stalemated was in earlier days considered a loss, and that the legendary World Champion Emanuel Lasker also considered the present rule to be idiotic.

Let’s subject the stalemate draw rule to a little informal analysis. Several sections in FIDE’s Laws of Chess contribute to the stalemate rule. The basic definition is in section 5.2.a, which reads in part:

“The game is drawn when the player to move has no legal move and his king is not in check. The game is said to end in ‘stalemate’.”

If one wishes to make Nigel happy (and who doesn’t?), there are at least four potential rule changes, some more interesting than others, that could achieve that noble end:

1. First and most straightforward (though least artistic): we could simply change rule 5.2.a to say that a player who is unable to make a legal move on his turn loses the game (whether the king is in check or not). Problem solved – a “stalemate” would now be a loss. But how satisfying is this?

To my mind, the “moral” argument that the stalemated player doesn’t deserve a draw runs up against the moral argument that a player who doesn’t achieve the aim of the game doesn’t deserve a win! The aim of the game – if we need a reminder – is to checkmate the opponent’s king. (“Checkmate” is really an announcement to the enemy king that he can’t avoid immediate capture. The custom of ending the game before the actual capture, I suspect, has something to do with ancient courtesy toward the monarch.) So in stalemate we haven’t really achieved our goal. We’ve marched our army to Babylon or wherever, and we’ve besieged the palace. The enemy king is ours, except that we don’t actually have him bound in our tent. He’s still untouched in his palace – we can hear the music and the dancing girls. Somehow that doesn’t sound like we’ve finished the job. We’re supposed to deliver an enemy king, dead or alive, and we haven’t got him. So have we won the battle where we sit mired in the mud, or have we simply blown it?

2. Another rule change that would avoid stalemate altogether would be to forbid a player from making a move that leaves the opponent without a legal move. This rule (according to Wikipedia) apparently occurs in some chess variants and has at times been enforced as a loss for the stalemating player (!), and it certainly prevents stalemate, but seems totally at odds with the aim of the game. Chess is a game with an aggressive purpose; bringing in a rule where the attacker must concern himself with providing his victim elbow room hardly deserves a hearing.

3. The third possible rule change is more meaty, and has interesting consequences beyond avoiding stalemate. This seems to be Nigel’s choice: he pointed out that we could solve the embarrassment to our game of having such a stupid rule as the stalemate draw (that’s not a direct quote, but I think I got his sense), and also experience the satisfaction of actually capturing the evil king by changing the rule that forbids the king from committing suicide. FIDE’s rule 1.2 says in part,

“Leaving one’s own king under attack, exposing one’s own king to attack and also ‘capturing’ the opponent’s king are not allowed.”

But why these prohibitions? What is gained by this rule? What is wrong with the idea that if a player is careless enough to put his king en prise, the opponent could take it and thus end the game, as in blitz chess? If we throw out this rule that forbids suicide, a stalemated king would have to make a move that places him en prise, whereupon he would be captured. (It is possible to construct a legal position where the king is entirely hemmed in by his own pieces, none of which have a legal move – in which case voiding the suicide clause wouldn’t help: we’d still have a stalemate; but such a position seems just theoretical – I’ve never seen it in a game.) Nigel is right: this is the rule that creates the dilemma of stalemate. Without this rule there is essentially no stalemate. So a simple way to make Nigel happy is to abandon this rule, and make the defending king self-immolate in a stalemate situation. Of course, the defender could simply resign instead, but the outcome would be the same: the “stalemated” side loses.

4. A fourth rule change that would alter the stalemate rule would be to allow a player to pass on his turn. The relevant rule is Article 1.1:

“… chess is played between two opponents who move their pieces alternately …”

So when it’s your move you must make a move. But why should passing not be allowed? What would be the effect of a player saying “I pass” on his turn? In most cases, this would merely lose a tempo and allow the opponent to improve his position. But there are cases, such as in a zugzwang or triangulation situation, or in play with the knights, where losing a turn may help a player. On the other hand, if one player can pass, so can the other, seemingly eliminating the advantage of passing.

In a stalemate situation one player has no legal move and would therefore have to pass. The opponent has a choice: he could pass to bring about the draw (three consecutive passes = draw, say), or he could make a move. For example, illustrating with a classical king and single pawn stalemate from Nigel Short’s own practice:

Sokolov-Short, Staunton Memorial Tournament, 2008

Short has just played 64…Kf3-g3, stalemating his opponent. White to play would have to play 65. pass, and Black would play 65…Kf3. If White is now allowed to pass again, we get nowhere. So for a pass rule to help Nigel out of the stalemate draw it has to be limited to positions where there is no legal move. In that case, after 65. pass Kf3, White would have to play 66.Kh2, and Nigel would win.

But the pass rule won’t help us out of all stalemate situations. For example, if the pawn is on a rook file, the black king’s sidestep (Kg3 in this case) does not give White a legal move, so the position would still be a draw (unless we invoke the suicide rule above). So allowing (actually requiring) a pass when there’s no legal move would allow the stronger side to win many stalemate situations, though perhaps not enough to make Nigel fully happy.

As we see, there are several ways to make Nigel partly happy, but what if these proposals make more chessplayers less happy? A utilitarian would attempt to sum and compare the resulting happiness and unhappiness to decide on the “best” course of action. I’m not convinced that happiness is additive, or even quantifiable, so I’ll try a qualitative approach.

For those who wish to make chess a simpler game, getting rid of the stalemate draw is a good way to go. Gone would be much of the subtlety of endgame theory, such as the “opposition”, triangulation, concern over “rook pawns” with the wrong bishop, and in general most pawn play. We may ask whether such complications add to or detract from the game.

For me, the answer is clear. The possibility of stalemate sets additional difficulties in the way of the stronger side on his way to a win. It gives an opportunity for the defender to overcome material and positional deficits, and to use his wits to give the stronger side a final opportunity to go wrong. Remove the stalemate draw and we have a materialistic game where the winning method would be to win a pawn and transition to the end game. This would impoverish the game as well as the world of chess composition, where many a beautiful idea has been spun around the stalemate theme.

The diagram is just one example of the surprising turn of events and the dangers that are still found in many positions even when the game is completely won. Here, Black to move wins easily, but he has one hurdle remaining: he must avoid stalemating White.

Carl Pilnick-Samuel Reshevsky, New York 1942 (US. Ch.)

Black to move

Reshevsky played 92…g4??, planning after 93.Qxg4 to win the pawn ending after 93…Qe1+ and Qg3+, exchanging queens. But Pilnick answered 92…g4 with 93.Qf2! and Reshevsky had to accept the draw. Did Reshevsky deserve a full point for blundering and failing to convert the win?

I find myself in agreement with the late GM Larry Evans, who called the suggestion to make stalemate a win for the stalemating player a "crude proposal that ... would radically alter centuries of tradition and make chess boring." (This Crazy World of Chess, cited in Wikipedia).

The question in the end is, should we change the aim of the game of chess? Instead of checkmate, should the aim be just to lay siege, to immobilize the enemy, as some propose? In boxing, you can win a match by knock-out (i.e., you’ve actually beaten the enemy) or on points (you’ve bothered the enemy more than he has bothered you). In chess, our tradition over some centuries has been that you have to beat the opponent to get the win. The suggestion to change that, to grant a win for cornering the opponent, for petering out in inaction while failing to find the coup de grâce, will take much of the complexity, subtlety, intuition, risk, and beauty out of the game, and with it interest for the game. The question deserves (perhaps) further discussion, but I see nothing but loss for the game if we follow this suggestion.

In part two, which will appear shortly, I will show you a number of practical and artistic cases
of stalemate. If you have any favourites yourself please send them in for possible inclusion.