There are a couple of issues with the definition of double attack in the glossary:
"Two attacks made with one move: these attacks may be made by the same piece (in which case it is a fork); or by different pieces, for example in a discovered attack when the moved piece also makes a threat."
The first issue is that this definition does not reflect the cited source, given by Yasser Seirawan in Winning Chess Strategy:[1]
"An attack against two pieces or pawns at the same time."
The more serious issue is that the glossary definition is too narrow. By defining a double attack as "two attacks made with one move", it excludes positions in which two attacks exist simultaneously even though only one new attack was created by the move itself. In other words, not every double attack arises from a move that independently creates two attacks.
A classical example occurred in a game between Emanuel Lasker and Max Euwe at the Nottingham 1936 chess tournament:
| a | b | c | d | e | f | g | h | |
| 8 | | 8 |
| 7 | 7 |
| 6 | 6 |
| 5 | 5 |
| 4 | 4 |
| 3 | 3 |
| 2 | 2 |
| 1 | 1 |
| a | b | c | d | e | f | g | h | |
White has just played 23.Kd3, attacking Black's knight, and Black replied with 23...Ba5, reaching the position shown in the diagram. White now plays 24.b4, blocking the bishop's attack while also attacking the bishop itself. The move creates only this single new attack, while White's king continues to attack Black's knight. As a result, after 24.b4 Black is subjected to a double attack against both the bishop and the knight.
According to the glossary definition, 24.b4 would not qualify as a double attack, because the move itself creates only one attack. Under Seirawan's definition, however, the move clearly does constitute a double attack, since two enemy pieces are attacked simultaneously, regardless of whether both attacks were newly created by the move.
Even Seirawan's definition is arguably incomplete, since it does not account for cases involving multiple simultaneous threats that are not direct attacks. However, that is a separate issue.
Erukx (talk) 02:41, 20 May 2026 (UTC) Erukx (talk) 02:41, 20 May 2026 (UTC)
- You're correct that Seirawan's definition is incomplete in the way that you suggest, but it's actually just wrong. Double attacks are attacks made with one move. Double attack describes the move and not the position, and a move that attacks only one piece or makes only one threat is a single attack irrespective of how many enemy pieces are en prise before the move. Hooper & Whyld have this definition: "a simultaneous attack against separate targets: these may be enemy men, or squares the enemy needs to defend. A double attack by one man against two is called a FORK. A double attack by two men could arise in several ways; e.g. from a DISCOVERED ATTACK." It's clear from those examples that 24.b4 is not a double attack and the term is never used that way in the chess literature. The threats are concurrent (existing in overlapping times) and not simultaneous (happening at the same instant). Simultaneous requires that both threats are created by a single move. Quale (talk) 02:58, 21 May 2026 (UTC)
- We should be getting our definitions from sources (Hooper & Whyld, or Seirawan) or common usage. In the tournament book, or other places where the game is annotated, is the situation in Lasker-Euwe commonly described as a "double attack"? If it is, then we should make sure our definition covers that, but if it isn't, that game isn't relevant for us.
- Also, I don't see why Seirawan's definition is "just wrong". It looks to me like it is compatible with Hooper & Whyld's definition. But our present definition looks more similar, word for word, to Hooper & Whyld's, so I wouldn't object if somebody wanted to cite Hooper & Whyld instead of Seirawan. Bruce leverett (talk) 13:18, 21 May 2026 (UTC)
- Seirawan's definition is wrong because it is too broad and also, as Erukx points out, too narrow. It's too broad because it includes any situation in which two pieces are en prise even if the original attacks were made 10 moves apart and not simultaneous. It's too narrow because it excludes double attacks in which a move makes simultaneous threats that don't have to be captures, for example a double attack move that threatens a piece and mate. Quale (talk) 02:41, 22 May 2026 (UTC)
- In Seirawan's definition, the phrase "in one move" rules out the situation where the original attacks were made 10 moves apart. Right?
- Seirawan's sentence is narrower than Hooper and Whyld's, in that Hooper and Whyld explicitly say "these may be enemy men, or squares the enemy needs to defend", whereas Seirawan says "pieces or pawns". Our current definition doesn't explicitly line up with either of those. If that's important, then we are in an awkward place. But if it's not important, then we can cite either one of those worthies. Bruce leverett (talk) 03:49, 22 May 2026 (UTC)
- I agree that one move can create two attacks simultaneously, but to claim that the position after 24.b4 in the game between Emanuel Lasker and Max Euwe, where two pieces are attacked at the same time, is not a double attack contradicts Yasser Seirawan's definiton. Nowhere in Seirawan's definition does it state that both attacks must be created simultaneously by a single move. When I said that Seirawan's definition of a double attack is incomplete, I meant that it does not cover all cases of double attacks, especially those in which at least one of the threats is not directed against a piece, but instead involves another tactical objective, such as the promotion of a pawn.
- Now consider the definition given by Hooper and Whyld:
- "A simultaneous attack against two separate targets; these may be enemy men or squares the enemy needs to defend, A double attack by one man against two is called a fork, A double attack by two men could arise in several ways, e.g. from a discovered check."[2]
- The definition clearly states that there must be two separate targets attacked simultaneously. It nowhere specifies, however, that both attacks must have been created by a single move. While the examples of fork and discovered attack do indeed create a double attack through one move, this does not imply that such cases are the only possible form of double attack. The definition plainly encompasses positions in which two simultaneous attacks exist, even if both were not created by a single move.
- It seems to me that the examples of fork and discovered attack where mistakenly incorporated into the glossary definition in a way that transformed double attack from a positional concept involving two simultaneous attacks into a move-based concept in which a singular move must create both attacks.
- Accordingly, the glossary definition limits the meaning of double attack in a way not supported by the cited sources.
- Erukx (talk) 09:37, 26 May 2026 (UTC)
