I hate the threefold repetition rule

The phrase "at least 75 moves" means that the draw does not apply for any such sequence less than 75 moves. The wording is, of course, unusual given that the rule itself dictates there can be no more than 75 moves, but that does not make it imprecise. The wording was no doubt taken from the same wording in the (non mandatory) 50 move rule stated in the preceding version of the laws.

If the arbiter doesn't terminate the game and the players continue making moves it doesn't matter. The moves after the expiration of the 75 move count are not part of the game. The game terminated in a draw at the expiration of the count.

I am using the term "legal move" as defined in art.3.10.1 in my previous post. I don't think interpretation comes into the issue. I don't have any complaints about imprecision regarding this particular issue.

I agree with MARattigans understanding of the 75M rule and I agree with Numquam that the formulation of the rule is stupid. I explained in an earlier post (#68) why I think it is the way it is but it will definitely confuse players and arbiters. Someone will get checkmated at 78M!

To resolve issues on the terms "legal" and "illegal" moves, we would first need to define what a "move" is. The handbook seems to indicate that a "move" consists of a chess action detailed in the articles 3.2 - 3.8 (something I call modular move). It means for instance that you cannot continue a 75M terminated game with a legal or illegal move because (formally) then there is no longer a board with chess pieces and no move can be made. In terms of a BBC game show by the same name and same concept: moving is IMPOSSIBLE. Similarly, placing a dragon on the chessboard is not a legal or illegal move; by the definition given, it is no move at all! It is however not certain that you can or cannot make a legal move in an illegal position. And can you castle in an illegal position?

You are assuming that moves made after the 75 move rule are not part of the game, but the rules do not say this. If the state has been reached that a player checkmates after 80 moves without capture or pawn move, then it is a win according to the rule. If you really need to make assumptions, then that means the rules are bad. The dead position rule is clearer than this rule for sure.

There is a normal interpretation of the phrase "legal move" relating to a game with moves. It means a move made in accordance with the rules. That being the case the phrase "legal move" cannot appear with its normal meaning within the rules, because then you cannot fully know what a legal move means until you know what a legal move is. Bertrand Russell would have claimed that a set of rules that used the term "legal move" in its normal sense was neither consistent nor inconsistent, but strictly  meaningless (See the discussion starting on page 37 here - https://ia800602.us.archive.org/35/items/PrincipiaMathematicaVolumeI/WhiteheadRussell-PrincipiaMathematicaVolumeI.pdf).

A set of rules may use the term "legal move" in the text so long as it  also contains a definition of the term (which may have little or nothing to do with its normal meaning). In that case any reference to the term should be interpreted only according to the definition. This is what the FIDE laws do.

The definition of "move" in the handbook makes no reference to termination of the game, so there can exist moves and also legal moves and sequences of legal moves following termination of the game, but the rules that result in termination of the game imply that these moves cannot actually be made by players as part of the game.

You cannot validly make a legal move in a an illegal position in a game played according to the rules of chess, because you cannot validly reach the illegal position. Again there is a normal meaning to the term "legal position" which is subject to the same considerations as "legal move". FIDE do define the term, but rather pointlessly, because the only place they actually use it they redefine it specifically for the rule in question. In any case they just define it to mean what you always thought it meant.

The definitions of "move" and "legal move" do not refer to "legal position" so could be applied to illegal positions outside of chess games, however that depends on how much history you assume for the illegal position, e.g. you can't castle if you assume a history in which the king has moved.

There is a point to MARattigans assumption that the >75 moves part was copied from the 50-moves rule but it is not an acceptable excuse because the situation is quite different. Are you checkmated after 50M, then it is indeed checkmate but not so after 75M. It's a gross error in rule design! There is probably a million of ways to write rules which are not formally imprecise, but only a few of them will be assimilated by chess players.

I don't agree. - art.9.6 states unambiguously that it's a draw if 9.6.2 (expiration under the 75 move rule) occurs. (See the handbook or post #85.) If the game is already drawn there can be no further play in the game.

The checkmate alluded to in 9.6.2 refers to a possible checkmate on exactly the last move of the 75 move period, which then takes precedence.

If the players don't notice the expiration and play on, then it's not part of the game any more. A checkmate is then not relevant. The game at that point is an ex-game; it has gone to meet its maker.

But the rule is still valid and not imprecise. It says that the draw will apply  whenever PC≥150. Since PC is monotonically increasing that will first occur when PC=150. Because the game is then drawn PC>150 will not occur. Obviously it would have been easier just to say the draw applies when PC=150. Somebody was clearly a very lazy typist.

Ah, I knew we'd end up there one day. I'm slightly ahead of you, I draw my formal language cues from Gödel

Yes, there are always context levels, especially when separating formal language from natural language. However we are now discussing different context levels applied to the chess domain and I am not too happy about unncessary distinctions. I'd rather have unified definitions of legal and illegal moves, legal and illegal positions and so on, such as to avoid speaking with two tongues. Since the handbook is the formal basis of organized chess, it is likely to be quoted on all levels of chess play and chess talk. Kind of unsatisfactory to have it mix with different definitions of terms living elsewhere in the community. I know I can write endlessly on this theme, like you can, but it will not get us anywhere concretely. I will just mention a few insights to show how important correct definitions in this area are:

• Retrograde chess consists mainly of retracting moves - more in particular illegal retraction moves, because wrong answers are easier to come by than right answers. But how can an illegal retraction be part of the content? How can one illegal retraction be separated from another one in a different category (e.g. returning all units to the PAS in a single sweep)? It requires legalizing certain illegal retraction moves. And that in turn requires defining modular retraction moves as a more basic entity than legal or illegal retraction moves. In other words, it requires separating the semantic values from the syntax of the valid retraction moves.
• The separation of syntax from semantics is equally necessary in forward chess to define an algorithmic approach to a posteriori logical problems (retrograde). To make analytical selections, otherwise illegal moves are temporarily legalized (as imaginary moves, the illegal variant of modular moves) in a context of first order logic.
• Definitions of checkmate and stalemate can be made by a combination a modular moves and a simple form of a posteriori logic and they are much more transparent and human than the current ones.

On the subject of 'moves' it is easy to mix up the abstractions with the instances of the moves. I avoided the distinction because on application there will always be an instance. For players, the possibility of a move does not exist outside the context of a running game so it is OK to declare it impossible. But I agree there are more ways to address questions of this nature.

But they can make such a sequence. The only requirement is that the moves are legal. They don't have to play the sequence while using the rules of a tournament game. The rules do not say that, that is your assumption. 

I have thought about and I came to the conclusion that the only viable interpretation is that the 'last move' in the rule, is the last move which was played. Otherwise this last move would depend on the sequence and depending on which sequence you choose the result could be a draw or a win. The rule applies to ANY sequence not just one specific sequence. So even if you take sequence 75-150 and white checkmates at move 160, then the result is a win for white because move 160 is the last move played.

9.6.2

any series of at least 75 moves have been made by each player without the movement of any pawn and without any capture. If the last move resulted in checkmate, that shall take precedence.

"The last move" clearly relates to the series of at least 75 moves made by the players without pawn moves or captures, and yes the last move played in that series (and the game). There will be at most one sequence  leading to the rule being applicable in any particular game because the game is then terminated, usually in a draw, but if the last move of the series is checkmate in a win. If Black's move 75 were the last pawn move or capture then the rule would become applicable on Black's move 150. The rule would then not terminate the game in a draw if Black's move 150 (the last move of the series and game) were a checkmate, but otherwise would declare the game drawn at that point. Either way the game terminates on Black's move 150 so there is no move 160.

The problem, I think, is the phrase "at least". The rule would have exactly the same effect if it were removed. The wording doesn't render the rule imprecise, but it's still pants.

The rule says ANY series, not just the specific series from 75-150 and no the game is terminated at move 160. The rule says 'the' last move, not the last move of the series, though you could construct a series such that move 160 is the last move. Imagine the arbiter looks at the game after all 160 moves have been made, then if he would follow the rule literally then white wins. The rule does not say that moves after 150 are not part of the game. If it does not say so, then they are part of the game. The game terminated at 160 with a checkmate.

Besides that the rule says at least 75 moves, it also says any series. So even if you remove the at least part, you can still construct multiple series: 75-150 and 85-160 in my example.

If you confound the FIDE definition of "legal move" with its natural meaning, then the players would be able to play out any series of legal moves, but then they couldn't construct suitable series in the first place in the situations under discussion.  

______________________________________________________________________________________

I "hate" people who "hate" this rule! Ever notice that it is usually real low rated players who dislike many of the rules of many games. To them I say move on to another game (or sport) or try to get better at whatever you are playing at. For those who call me "low rated", my rating is on this site`s blitz Chess of which I admit I suck at. My USCF Correspondence rating is 2120 and my ICCF rating is 2219 and i was once 1764 in the USCF OTB ratings. My point is that I`m a fairly decent player and I know what the Hell I`m talking about!

@Numquam

Out of interest, prior to 2017 the dead position rule was overridden in the Competition Rules section by

which appears to be exactly the way you want the rule to work. The overriding article was excised in 2017.

I never hated this rule but I am on the verge of hating it now. So much brainwork involved

Some notes with additional views:

Where tournament (competition) rules do not apply, the 75M rule does not apply and there is no issue.

There should not be an issue with "legal moves" in this matter as there is a priority condition preceding the "legal moves". Before a "legal move" can be played, one should first establish that the game is not terminated by something that occurred earlier. The "right to play" determines whether or not subsequent "legal" or "illegal" moves can be considered at all. It is a tail end evaluation of the preceding move and not of the next one.

In general, the outcome of analytical moves is not different from actual moves for the following reason: there must always be a proof game. By stating that a particular series of "legal moves" is available, you must be able to produce a proof game with real moves for that series which does not transgress any rules. It is inevitable that when such a proof game passes the 75M line, a draw is assigned and the game is terminated. After it, the "right to play" has disappeared and nothing can happen. If you claim a position is not dead at some point, you must be capable of producing a legal proof game with a win/lose outcome.

It is valid to permit different requirements for "legal moves" (and "right to play") in an analytical move series from a real move series, but such should be specified by the rules. This is for instance the case when analyzing checkmate. The analytical actions to justify a checkmate verdict look very much like normal "legal moves" but of course they violate several rules. That is why FIDE chose to circumvent the common concept of "legal move" at this point. Consequently, you need not present a proof game for capturing the opposing king.

Hope this helps in the discussion.

Nice find. The definition of a 'series of moves' is missing in the handbook, but we all know what is meant. When reading rules I try to make as few assumptions as possible. What you are saying is that a 'series of moves' depends on if it is a game with or without tournament rules. I don't agree with that. I use only article one. The rule dead draw is the same for both informal games and tournament games. No clear distinction has been made in the rules.

And btw that position with white to move is illegal, because the game ended due to stalemate. Even if it was legal then it isn't dead because black can get selfmated.

Edit: I responded to a deleted post.

@Numquam

I realised the position I gave was not dead anyway so I'd deleted it before you answered. Give me a mo. Corrected version hopefully on its way.

OK: I don't think the change of position affects what you said apart  from the last two paragraphs.

You would presumably also argue that the following White to play position is not dead.

The FIDE definition of legal move requires only that arts.3.1-3.9 be met. Those do not include

1.2 The player with the light-coloured pieces (White) makes the first move, then the players move alternately, with the player with the dark-coloured pieces (Black) making the next move.

so Nc8, Ne7, Ng8# is a series of legal moves resulting in checkmate, but the series could not be played because of art.1.2.

That would be the case, at any rate, if FIDE had added a rule stating that a player may make a move with only one of his own pieces, but that omission is clearly an oversight.