Neglecting to remove pawn captured en passant

I read about an interesting situation in a blog post the other day. After White played a2-a4, Black played b4xa3 but neglected to remove the a4 pawn before pressing the clock. White then quickly made another move in reply, continuing to leave the a4 pawn there, and Black got confused about what had happened and just continued the game from there, and nobody ever removed the pawn.

I assume that capturing en passant but not removing the captured pawn before hitting the clock is formally an illegal move? If White notices that the move is illegal but the position on the board favors him, is he required to note that the move was illegal? The phrasing in the 4th edition rulebook (that’s all I have) all says things like “If it is found [that a move was illegal]” without mentioning the circumstances of the finding.

Should White have

  • just kept playing, as he did,
  • removed his a4 pawn because that’s clearly what Black meant to do,
  • stopped the clocks and told Black that he made an illegal move,
  • summoned the TD?
    (I guess the last response is always appropriate.)

Also, the rules say that a spectator may not point out illegal moves except to the TD, and that the TD has the responsibility to correct illegal moves. Does that mean that a spectator can/should go get the TD to come correct the position?

Black did not play b4 x a3 in this example, black simply diagonally moved his pawn. Spectators are just that, spectators. Spectators should not summon directors, one of the players in the game should summon a director if and when they need help.

(b) might work out OK, but takes a slight risk of creating a scene. (c) and (d) together would be a good alternative. (a) is the worst answer of the four.

As for a spectator pointing out an illegality privately to a TD, I don’t think that’s so bad. That action does, however, put the TD in the uncomfortable position of having to choose whether to take action or ignore it.

Bill Smythe

Tournament directors can’t be everywhere at once. Why not give advance notice before the event that spectators are “empowered” to be director’s assistants? I would also warn my assistants that spurious claims by them would result in ejection from the tournament site.
This would not be totally out of line because I saw a scholastic tournament being run where the adjudicator could not play chess. He counted white units and then he counted black units, awarding victory to the player with “the mostest.” Thankfully, this was not a rated event. Choice B would be the best choice if the player intended to play p x p. Either the opponent would object and cause an immediate crisis requiring the director or they would not and life would go on.

I think this is a very bad idea. I also think having an incompetent adjudicator is a very bad idea.

Alex Relyea

Replacing the how-to-resolve answer that was eaten by the website outage (not complaining; it happens):

Assuming USCF rules:

Black has determined the move bxa3 e.p. by releasing the pawn on a3. Black has completed an illegal move by pressing the clock without removing the white pawn on a4.

If a director observes the illegal move he should initiate correction himself (unless the tournament follows a strict non-interventionist policy–this is an exception, but one that is used at a number of major USCF scholastic tournaments). If it is not observed by a director, the players have a certain number of moves within which an illegal move may be pointed out and corrected (usually ten, but two in blitz).

If noted within the number of moves allowed for a correction, the position reverts to the last legal position, and White is awarded two minutes for Black’s illegal move. Black is compelled to play bxa3 e.p., because that move was determined when Black released the pawn on a3. No further time adjustments are made; the time elapsed untill the call of the move is lost.

If the illegal move is not pointed out until after the number of moves for a correction has passed, the illegal move stands, along with any subsequent moves, captures, or promotions by either a-pawn involved.

In most cases, a spectator intervening will be directed to leave (and I have escorted a spectator out of the playing hall of the Eastern Open before).

I haven’t thought this through enough to agree or disagree forcefully, but can you cite a rule from the rulebook that supports this position?

Alex Relyea

This is an interesting problem. It is an illegal move either because the pawn was not removed or because the pawn was played to the wrong square like moving a Knight from g1 to g3. It is a pet peeve of mine that promoting a pawn without exchanging it for a legal piece is not treated as the illegal move it clearly is and this case may be similar. Correcting the problem depends upon the illegal move variation being used, the time left (going back either 10 or 2 moves), and the claim being made. Touch move would apply and if there were no pawn on the rank in front, either pxp ep or pawn forward one square could be played as by not removing the pawn it is unclear what pawn move was intended.
Regards, Ernie

Was this before the advent of ubiquitous smartphones?

There is an official free version of Stockfish available for iOS devices, and an unofficial version available for Android devices, so it should be no problem nowadays to use a strong computer to adjudicate, even in an unrated scholastic tournament.

–Tim Smith

The tournament was a unrated scholastic event run in a manner to end the tournament quickly, actual chess be d#@$. This was certainly before the smartphone period but if this particular group were still running events (and I sincerely hope that they are not) I believe they would stick with their efficient, tried and true tournament shortening method.

That’s not correct, is it?

This rule, along with others governing the determination of a move, needs a lot of work, but I don’t see how releasing the pawn on a3 constitutes determination of the move by the above definition.

Bill Smythe

I agree. The Club TD certification easy enough to obtain. If a person cannot be bothered to buy the book, read it, and attest to the fact that he has done so and will abide by those rules, then I really don’t want him in the role of Floor TD.

Players pay their dues and entry fees with the expectation of having certified directors in rated tournaments.

The rules on move determination don’t specifically address en passant, but I think that by applying similar situations that are specifically addressed, one would reach the conclusion that the move has been determined.

In the case of translation moves, the move is determined when the piece is released. Having moved the pawn from b4 to a3 and releasing it, the player is obligated to move from b4 to a3 if it is legal to do so. The only way to do so is to capture the pawn on a4. Therefore, I would say that the move bxa3 e.p. is determined when the pawn is released on a3.

ETA: and if anyone read an earlier version, I corrected my notation. I think it ought to be bxa4 e.p., but that is apparently not the convention.

The convention is based on the history of the evolution of the move - which is why its bxa3. (It used to be 2 moves as Pawns could only move 1 square at a time - hence when the game was speeded up by allowing pawns to move 2, an accommodation was made to allow for the one-time possibility of a capture - in this instance at a3 - which would have otherwise been lost. Hence, bxa3.)

Rule 9A. An en passant capture is both a capture (9B) and a transfer to a vacant square (9A). Pursuant to 9A, the move is determined when the hand quits the piece on the vacant square.

I do not think that is correct for Blitz:

I am guessing that Mr. Price meant to write “two if there is time pressure” rather than “two in blitz.” It is, of course, correct that under both USCF and FIDE blitz rules, the opponent may claim a win after the player completes an illegal move. (To be pedantic, I assume that in the case of FIDE rules, there is “inadequate supervision,” meaning there is not one arbiter per game. If there is, then the illegal move is treated as it would be in regular play.)

That depends whether you mean determined (lowercase) or Determined (uppercase). Sort of like the distinction between democratic and Democratic.

It might be reasonable to define a move as determined (lowercase, generic) when it has reached the point where, due to touch-move and other rules, there is only one legal way to complete the move.

However, the rulebook spells out specific definitions of Determined for several different cases – transfer, capture, promotion, castling, etc – some of which differ from the generic notion. Unfortunately, though, the rulebook uses lowercase in these specific definitions, thus muddying the distinction between the generic and the specific.

Further muddying occurs with the use of the phrase “determined with no possibility of change”.

The confusion between determined and Determined has some strange side effects. But that’s a whole new topic. I think I’ll start a new thread in a day or two.

Bill Smythe

Agreed. There’s a certain fuzziness.

As a TD, though, I would treat Determination of en passant as a missing case in the rulebook, and apply some extensions of existing rules. My interpretation would be that a pawn moving diagonally to an unoccupied square Determines an en passant capture.

And it does create an interesting dilemma. Is moving diagonally, but failing to take the pawn, an illegal move, or is it just an incomplete move? In other words, if it is determined with no possibility of change, then he has made a perfectly legal move. He just didn’t execute it correctly.

Likewise, if a king moves two squares, and then the player punches the clock without moving the rook, has he executed an illegal move, or has he committed to a perfectly legal castle move? The difference is that in the former case, the opponent gets a time adjustment, but in the latter case someone says “j’adoube” and moves the rook, and carries on.

ETA: And I wouldn’t actually capitalize Determined. I would provide a specific, technical definition at the beginning of the chapter, and only use it that way in the chapter. But your use of caps illustrates the problem.

I’d have to say both examples should be treated as illegal moves, and there should be no such thing as an incomplete move if the clock has been hit. Once you hit the clock, you give up all rights to claim an incomplete move, so saying “j’adoube” should not allow you to move the rook to a new square after you’ve hit the clock.

Basically, as long as you haven’t hit the clock, the move is incomplete. Once you hit the clock, the move is illegal. That’s how I see it anyway.