Three-fold repetition and en-passant

@Lagomorph It seems that you're only saying you don't like the written style of the rule, without showing it's actually ambiguous.

In 99.99% of cases, positions with exactly the same pieces on the same squares would also have the same possible moves. The only exceptions are situations involving e.p. and castling and that is why (1) 9.2.2 mentions that the pieces must have the same possible moves too to cover these unusual situations, and (2) the sub-rules about e.p. and castling exist to highlight and elaborate on these special cases.

Since 9.2.2 is about all possible types of moves (capturing and non-capturing) and 9.2.2.1 is about e.p. captures in particular, how can you expect their wordings to be exactly the same? You even say that the different wordings amount to the same thing, which only proves that 9.2.2.1 and 9.2.2 are consistent, not ambiguous.

To mention e.p. and castling specifically in 9.2.2. would be to overload that rule with information. It's much clearer to have these separate sub-rules to elaborate on the main rule. Again, this is a matter of style, not ambiguity. 

He didn't "agree with me". I asked how this situation should be interpreted, and he presented his opinion. I did not present any sort of interpretation myself at any point. I'm just repeating what he wrote.

Here is what he wrote:

Thanks for quoting the email, DjVortex. Wonders never cease as the distinction between "possible" and "allowed" e.p. simply doesn't exist in the FIDE rules and a FIDE Arbiter would pull that distinction out of thin air.

His "argument" by analogy with Article 3.9.1 doesn't work either:

"3.9.1 The king is said to be ‘in check’ if it is attacked by one or more of the opponent's pieces, even if such pieces are constrained from moving to the square occupied by the king because they would then leave or place their own king in check."

A pinned piece could "dynamically" attack the opponent's K and place a real check, but that's only because in chess attacking a king is paramount, and thus here the pin is "weak" and not absolute. But if the pinned piece is attacking not a K but just an ordinary piece, the pin IS absolute and there's no "dynamic" possibility of making the capture. And guess what? In e.p. it's just an ordinary P being attacked and if the would-be capturing P is pinned, that pin is absolute and there's no "dynamic" possibility of making the e.p. capture either. Therefore the e.p. is simply "disallowed" at first by the pin and "disallowed" again later because it's too late, i.e. the position IS repeated.

It would seem that the Honorary Chairman of the FIDE Arbiters committee has proved my case for me.

Thank you DjVortex for publishing the text of his email.

That email only proves that even a FIDE Arbiter is human and could misunderstand a clearly written rule. I've yet to hear a legit reason (that isn't a non sequitur) for claiming that the 3-fold rule is ambiguous. People could mistakenly claim that 1+1=3, but that wouldn't prove that "1+1=2" is an ambiguous statement.

If you re-read the posts I don't think anyone is suggesting castling is "legal" in that position.

The point is that FIDE's rules make a distinction between being able to castle legally on any given move, and the "right" to castle at some point later in the game.

This is the whole crux of the argument in this thread.

Rocky64 wrote:

I think this might be a case where the intent of the people who wrote the text does not match the actual text, when the text is read literally.

While the text of the rules might not explicitly say it, I believe that the intent is to consider "right-to-en-passant-capture" to be a sort of "state" in which the game currently is, regardless of whether the capture can actually be done or not. That "right-to-en-passant" only looks if the previous move was a pawn moving two squares to the side of an opposing pawn, and doesn't care about anything else. This move creates a temporary "right-to-special-capture" that's in effect in the next move. This "state", if you will, exists even if the capture is illegal due to something else (such as the player's, whose turn it is, king being in check by something else than the pawn that just moved).

As you note, the rules text doesn't establish this, and instead use a wording that seems to indicate that if the capture is not possible, then it's not considered a possible move among the list of all possible moves, period. However, is that the original intent of the text, or is it just an oversight?

The interpretation of that FIDE arbiter seems to suggest that, at least in his opinion, the intent is to consider "en-passant rights" to be a "state" of the game that doesn't care if it actually can be done.

Personally, I saw some convincing examples and cases all over the thread and almost changed my stance on the matter, but then I was like nope. It's all about possible moves. Once that changes, the position changes.

Even the theoretically possible moves. Once they change, the positions holds a different value.

But why would the rule makers want to make that kind of distinction between e.p. "rights" and its legality in the first place? Why would they want to pretend that two positions are different when these positions are exactly identical with the same available moves? It's simply illogical and the idea that it was an oversight implausible, especially when you consider why the e.p. and castling sub-rules were added to the 3-fold repetition rule.

These sub-rules didn't exist in the earliest FIDE rules - that was the real oversight -  but was added when people realised two seemingly identical positions could have different possible moves, due to the special properties of the e.p. and castling moves. These positions are thus not really the same and they have real, practical differences between them (e.g. castling rights lost). Therefore it was worth complicating the 3-fold rule to take these special situations into account. Now, there's no real, practical difference between two identical positions where an e.p. capture is illegal in both. Thus it wouldn't have been worthwhile to complicate the rules further - making the distinction between e.p. "rights" and its legality - just to pretend these positions are different for the sake of the 3-fold rule.

I think that the FIDE arbiter that responded to me was thinking, as he said, of it the same as with a piece checking the opposite king doing so even if said piece cannot legally move. In a similar manner, the pawn has the "right to capture en-passant" even if it cannot legally move, and thus it is, in principle, part of the list of "possible moves" in this sense.

I already explained how the FIDE Arbiter's analogy is faulty in post#65. If you disagree, you should address my argument directly, before hypothesising that the rule makers omitted such an analogy in the rules by accident. Until you could show that the analogy is valid, the far more obvious reason why they omitted it is that they agreed it was faulty.

Introductorily (clears throat), I'd like to begin on an evocative note (stares awkwardly). There are two types of draws in the world of chess; a positional draw where no decisive progress can be made, and a conceptual draw where no progress can be seen by the player. So state the rules.

Rules constitute structure, certainty. They're the reasons my fingers do not break when I move them. If there is uncertainty surrounding the rules, then we may as well be walking on quicksand.

Ladies and gentlemen, bottom line is if there is a difference in a position (technical or theoretical), then that position is different. That is as clear to me as daylight. But in all fairness, I'd like for us all to participate in a little thought experiment that tackles a sensitive angle regarding the Three Fold Repetition rule.

Assume that you're in a tournament, a blitz tournament. It's your second game for the day, and you have 15 seconds left on the clock, your opponent has more. It's painfully clear that both players are drawn and the position has repeated a few times. You raise your hand to call the arbiter, and plan to claim a draw by rule, but the arbiter says that there are slight theoretical or technical differences in the positions you noted, and that even though the game is obviously drawn and repetitive, you must continue to play on until 3 truly qualified repetitions occur. Without much choice, you play on and lose on time, with the draw just 3 seconds away.

How would you feel?

After much debate (clears throat), I've decided to propose a technical extension of the Three Fold Repetition rule, to remedy the uncertainty regarding this rule, as I believe that no rule should ever be haunted by uncertainty.

The new extension should go as follows:
Considering that the purpose of the Three Fold Repetition rule is to remedy a situation where no further progress can be achieved by either side, it should be so that in the event of there being 3 occurrences of an identical position in a single game, but having these occurrences marked by slight technical differences in regards to each other, that these occurrences (in the relative sense) be deemed unfit and unqualified to serve as prerequisites to any claims regarding a draw. With the exception being the purpose of the Three Fold Repetition rule itself. Therefore, if in any single game the situation is as described above, but the game itself is indeed locked in a state in which no progress can be made and there are positions reoccurring (regardless of the slight technical differences), then a fourth candidate, of even a fifth candidate (a candidate itself being the positional recurrence [with or without any slight technical difference]), be considered the deciding factor and proof to the purpose of the Three Fold Repetition rule. With the second exception being that the slight technical differences between these candidates should not accumulate to a situation in which progress can then be achieved, as this would then defeat the purpose of the Three Fold Repetition rule, and thus the purpose of this proposed extension.

In simpler terms, if the recurring positions look the same but their identicalness is somewhat debatable, then one should look to see if indeed no progress can be made in the game, and if so, then consider a 4th, or even a 5th, repetition as the decider, or determinant.

So if these shoddy positions repeat 4 times, or even 5 times, it's clear that the game is going nowhere, regardless of those positions supposedly having slight technical differences. This is, of course, only true if said positions don't accumulate to something game changing every time they occur.

With all that said, I believe the proposed extension to be optional. That is because the 3 Fold Repitition rule itself, as currently is, should suffice in such situations. If indeed in any single game further progress can no longer be made, there'll eventually arise 3 repetitions indisputable. But as seen in the experiment above, the optional extension does have its applications where time is crucial.

Personally ?....I find much of FIDE rules, and more so USCF Rules ?, to be somewhat ambiguous....and left up more to the TD/Arbiter than anything.

But I can kinda see why that is....as a rulebook could end up like a NY phonebook w/ all the nitpick-nitwits that wanna get wrapped around the axle over subatomic minutia.

Long story short, the answer to the original question is that the 3rd occurrence would be repetition, unlike the first case where it took 4 because of the en passant option.

The rule is that it must be the same position with the same player to move, both players having the same legal options.  En Passant is NOT AN OPTION on the first occurrence because the pawn is pinned to the King.  EN PASSANT IS NOT A LEGAL MOVE.  Therefore, the first occurrence Black has the same legal options as the second and third, and White's legal moves are of course the same too.

If the pawn were pinned to anything other than the King, it would be different.  Let's say it was pinned to the Queen.  Then En Passant would be STUPID BUT LEGAL and what matters is whether it is legal, not whether it is intelligent!  So in that scenario, a fourth scenario would be necessary because en passant is legal the first time, but here, pinned to the King, en passant is not a legal move!

Lol

Hi five, Lola!