The Principle of Two Weaknesses
One weakness can usually be defended forever: the defender posts a piece, ties his king to the target, and waits. The principle of two weaknesses says that winning technique consists of creating a second target on the other side of the board, because a defense stretched between two fronts must eventually abandon one of them. It is the closest thing endgame play has to a universal conversion recipe — when strong players "grind," this is what they are doing.
Why One Weakness Is Never Enough
Defense is geometrically easier than attack. A weak pawn can be attacked from only a few squares, and the defender — who knows exactly what must be held — parks his king on the shortest path between the threat and the target. As long as there is one target, his resources concentrate; every attacking regrouping is answered by a one-move adjustment. The position becomes a fortress in the practical sense: not unbreakable by rule, but unbreakable by any sequence the attacker can actually construct.
The attacker's resources, however, usually include something the defender's do not: mobility to spare. An extra pawn, a more active king, a piece with free range — these surpluses cannot break a single point of resistance, but they can threaten two points in alternation. The defender, tied to front A, cannot arrive at front B in time; tied to B, he loses A. The win never happens at the weakness you have been attacking all game — it happens at the one created while the defender was busy.
Distance is the whole mechanism, which is why the principle lives on concrete squares. Here it is in its minimal form — six men.
Two Fronts in Miniature
White: king e5, pawns b4 and g4. Black: king c6, pawns b5 and g5. Both black pawns are fixed on attackable squares, and they stand five files apart — farther than one king can patrol. Black's king guards the b5-front. White to move wins by opening the second front:
1.Kf5! The g5-pawn is attacked, and no defense exists — from c6 the black king cannot reach a defending square in one move. Note the tablebase's verdict on the alternative plan: every move that keeps the kingside raid alive wins, while 1.Kd4?, committing the king to the front Black already holds, only draws. Black's best is the counterattack 1...Kd5, and after 2.Kxg5 Ke5 he shadows the white king, hoping to blockade the new passer. It doesn't hold: 3.Kg6 Kd6 4.g5 Ke6 5.Kh6! — the king steps aside to let the pawn through while covering g7 — 5...Kd5 6.g6 Kc4 7.g7 Kb3, and Black's counter-race arrives hopelessly late: White promotes, and the new queen wins the b-pawn race with checks to spare.
Play through the line again and watch Black's king. It spends the whole game in between — never quite defending the kingside, never quite winning the queenside. That torn, too-late feeling is what the principle of two weaknesses inflicts, and it is the sensation you are trying to create when you convert an advantage.
The Same Position Minus One Front
Now the control experiment. Remove both kingside pawns — same white king on e5, same "weak" pawn on b5, same defending king. The tablebase calls it a dead draw:
White can maneuver forever — 1.Kd4 Kb6 2.Kc3 Ka6 3.Kb2 Kb6 — and never make progress, because the b5-pawn can only be attacked from squares the black king always reaches first. One attacker, one defender, one target: arithmetic that never changes. The entire difference between this draw and the win above is the existence of a second target. Not a bigger material edge, not a better king — a second front.
This pair of positions is the principle stripped to its skeleton, and it scales up intact. In rook endgames the two fronts might be a weak pawn and an exposed king; in minor-piece endgames, a bad bishop's pawns and an entry square for the enemy king. The material changes; the geometry — two targets, one stretched defense — does not.
Where Second Weaknesses Come From
Nobody hands you the second front; converting players manufacture it. The standard sources, roughly in the order I check them at the board:
Fix pawns on vulnerable squares. A weakness must hold still to be a weakness. Advance your pawns to lock the opponent's on squares your pieces attack — a pawn that can still push or trade is not yet a target. This preparatory phase rewards patience; it is half of what do not hurry is about.
Use a passed pawn as weakness number one. A passer you may never promote still forces the defending king or a piece to watch its path permanently. The outside passed pawn is the purest case: its entire value is that it is a front the defender cannot ignore, opening the rest of the board for your king.
Treat king-entry squares as targets. The second weakness does not have to be a pawn. A hole your king can walk through — c5, or f5 — functions exactly like a target: the defender must guard the square itself, with the same stretching effect.
Probe both wings before committing. Play a threat on one side, watch the defense reorganize, then switch. Each switch costs the defender more time than it costs you — your pieces are more mobile; that asymmetry is your entire income. When a switch catches the defense mid-regrouping, that is the moment to strike, and the strike should usually arrive as zugzwang: the defender, stretched to his last tempo, must move something that holds.
Planning in these terms — targets and defensive burdens rather than move sequences — is the mindset shift covered under schematic thinking, and it sits at the center of the whole conversion toolkit mapped in endgame principles. When your winning attempts stall, the diagnosis is almost always the same: you have been pressing one weakness. The prescription is never "press harder."
Frequently Asked Questions
What is the principle of two weaknesses in chess?
It is the rule that one weakness can usually be defended indefinitely, so the winning side must create a second target on another part of the board. Alternating threats between the two fronts stretches the defense until it fails at one of them — usually the newer one.
Why can a single weakness be defended forever?
Because defense concentrates while attack must approach. A lone weak pawn can be attacked from only a few squares, and the defending king sits on the shortest path, answering every regrouping in one move. Without a second target, the attacker's extra mobility has nothing to stretch.
Does the principle only apply to weak pawns?
No. A "weakness" is any point that demands permanent attention: a weak pawn, the path of a passed pawn, an entry square for the attacking king, or an exposed king itself. Any two such burdens far enough apart create the stretching effect.
How far apart do the two weaknesses need to be?
Far enough that the defending king or pieces cannot cover both within the time a threat takes to execute — in king endgames, usually three or more files. Two weaknesses close together can be defended as one front and lose most of their value.
Is the principle of two weaknesses only an endgame idea?
It is most visible in endgames, where defenses are thin and the king must serve as a defender, but the same logic drives middlegame play on two wings — for example attacking the king while tying pieces to a weak queenside pawn. Endgame study is simply where the mechanism is easiest to learn in pure form.