Castling - Rook touched first

Running Chess Tournaments
1

I’m sure that this has been covered before, but I recently had occasion to read the rules closely, and they seem somewhat contradictory. It’s one of those things that I thought I knew until I read the rule book.

(References below refer to the printed 5th edition rule book.)

Rule 8A2 describes the procedure for castling. Move the king two squares, then move the rook. However, it refers the reader to 10I2, in which the rook is touched first.

Meanwhile, rule 9A describes the rules for when a move is determined. If a piece is transferred to a vacant square, and then released, the move is determined with no possibility of change.

Suppose white has the intention to castle. He picks up the rook on h1, moves it to f1, and lets go. According to 9A, he has, at that point, executed a transfer to a legal square, and let go of the piece. It should be determined with no possibility of change. Furthermore, he was not following the procedure described in 8A2 for castling.

However, if that is the case, then how can we make sense of the existence of 10I2? The sensible way to touch the rook first while castling would be to move it to f1, then pick up the king and move it to g1, but that runs afoul of rules 9A and 8A2.

Surely the rulebook is not suggesting that one might touch the rook on h1, leave it positioned on h1, then pick up the king and move it to g1, then pick up the rook and move it to f1. It would be silly to specifically call for that move in the rulebook. Even sillier would be to pick up the rook on h1, put it down in an illegal position (like off the board, perhaps) then move the king to g1, then grab the rook and move it to f1. Silly or not, that is the only way I can make sense of the rules as written.

What was the actual intent?

2

9A refers to 7A,B,C for clarification on what constitutes a move. And 7A starts with: “With the exception of castling…”

So I would say that moving the rook to f1, when castling is the intent (ie, the player then touches the King and moves it to g1), 9A doesn’t come into play. When only the rook has been touched, placed on a different square, and released (as part of a castling maneuver), it is not considered a “move” (per 7A).

3

But if that is the case, then there is a situation where moving the move may or may not be determined. If it is a transfer, then the move has been determined. If it is a castling move, then it has not yet been determined. The whole point of “determined”, is that it cannot be altered. Perhaps one could say that the movement of rook to f1 has been determined, but it has not yet been determined whether or not the king would move to g1.

Since draw claims are to be made after a move has been determined, how would one indicate that the intended move as a transfer instead of a castle? By marking it on the scoresheet, perhaps? (Or one could go with the verbal analogy and say of another famous rule and say, “I castle” when touching the rook, except I would prefer “Je roque.”

As an aside, the immediate cause of this question concerns a Local TD exam I recently took. I don’t believe that any of the four answers are correct, but I can’t say exactly what is wrong with the answer that was asserted as correct, because to do so I would have to give the exact wording of the question. (I got the question right, so I know what is supposed to be the right answer, but I don’t actually think it is right based on the wording in the rule book. It’s just the one that is least wrong.)

4

This is an interesting line of discussion to me; I gave an answer as I am but a lowly club TD who is looking to take the local exam next year, and I wanted to see if my understanding of the rules matched that of the experienced and senior TDs. Your point seems to be that the rook move is in a kind of Schrödinger state of being determined and not determined, based on the intent of the player - is it part of the castle move or is it an actual rook move. Unable to say, without inspecting the mind of the other player.

Can you come up with an actual example where it would matter? Would an opponent yell out “draw!” when a rook moves to f1, thinking that was the determined move, and the first player says - “nope, I’m castling”. I’m trying to think of a position (with historical moves), where that could ever come up, and it would matter (whether the move was Rf1 or 0-0). Doesn’t mean it isn’t a fuzzy area of the rules, just wondering if it could ever arise in practice.

5

Interesting perspective. I think it would take a very contrived position to create a situation where Rf1 was a stalemate, but o-o was a victory for one of the players. It could probably be done, but not realistically.

Where it could easily arise, however, is exactly the circumstances that led to the adoption of the rule as it is. In FIDE rules, and in the most common understanding of touch moves, the king must be touched first. I believe, but am not certain, that it was so in earlier editions of USCF rules as well. This had the effect of a player (frequently a child) touching a rook with the intention to castle, and being informed that he may not do so. There are lots of cases where castling is a good move, but Rf1 is a very bad move. If a move is determined with no possibility of change (per rule 9A) as soon as a legal transfer move has been made and the hand left the piece, then someone could be forced to make a bad move that he did not intend to make.

That, after all, is the practical effect of the touch move rule. You really don’t want to make a move, but you must.

If I simply apply the English language to the rules, as written, it would mean that I can touch the pieces in any order. However, if I actually move the rook to f1 and then release it on the f1 square, that determines the move, and I must make that move. The rulebook says that in the case of castling, you can touch the rook first, but it doesn’t say you may move the rook first. In fact, it describes a procedure in which the rook is moved second as the proper way to castle. Taken literally, the rules say that if you move the rook to f1, that’s your move.

However, I have been told repeatedly, in person and via web pages, that in USCF rules, you can move the rook first, and then continue by moving the king. I’m pretty sure that that is the intent, and I can see no benefit in having rule 10I2 unless that is the intent.

As for the TD exam, I can’t discuss the specifics of the question. It wasn’t a simple, “Is it legal to castle like this” question, but had a subtle rules issue related to that question, and based on my understanding of the rules, none of the multiple choice answers were correct. That either means my understanding of the rules is wrong, or the question is badly worded.

6

The “intent” of the rule 10I2 is an unsavory appeasement for obstinate players who refuse to touch their king first when castling.

Appeasements create problems. For example, suppose the board has the following pieces:

  • White has: e1-K, h1-R, h2-Q (and White still has the right to castle short).
  • Black has: b8-k, c7-q.
    White is on turn, and his clock is ticking.

Here is the sequence of events:

  1. White should move Q:c7 check; but instead he unwisely decides to castle short.
  2. The obstinate White player begins to castle by first lackadaisically sliding his rook and releasing it on f1.
  3. With clever anticipation, very quickly after the release on f1, Black hurries his queen to take occupancy of h2, knocking the White queen off the board.
  4. Only now, after the capture and removal of his queen, does White clearly touch his king and thereafter move it to g1 attempting to complete his castling.

By chance the arbiter happened to see the whole event. Arguments by the players ensue.

White argues that his own intention was to castle, therefore Black must let White castle.
Black replies that there is no such thing as mind reading, and that the concept of intention in this context is non-verifiable gibberish.

White next argues that Black cannot reply with a move until after White has pressed White’s clock.
Black directs the arbiter to ChessCafe.com and to Geurt Gijssen’s column dated April 2012 (http://www.chesscafe.com/text/geurt168.pdf). In the column Geurt concludes that Black is allowed to move before White presses White’s clock.

The arbiter contemplates, and then passes judgment in favor one of the players. In the other player the decision causes either justified or unjustified outrage.

I dislike appeasement rules like this 10I2.
I also dislike the appeasement rule that allows a player to take limited notes by erasing his prematurely written move and writing a different move. This appeasement was voted in around 2006-8 I think, and it allowed no more than one erase+rewrite per move; meaning perhaps 60 rewrites per game per player. But a year or two later I read a rumor that the appeasement was dialed back somewhat to now allow only occasional rewrites?
. .

7

Utter nonsense. At the 2006 Delegates’ Meeting, rule 15A was modified to disallow recording the move before making the move on the board. Specifically, the last sentence of rule 15A was changed from:

to:

While this change did bring the USCF Official Rules of Chess into compliance with the FIDE Laws of Chess, the change met with considerable resistance. As a result, a variation was added at the 2007 Delegates’ Meeting that allows players using a paper scoresheet to record the move before executing the move on the board. Distressingly many players (and, regrettably, tournament directors) mistakenly believe this is the actual rule, not a variation. Of course, it is a minor variation and does not need to be announced in all pre-tournament publicity; nonetheless, it is still a variation.

There was never any “rule” that allowed one erasure per move. Please do not make stuff up.

It is also (in my opinion) incorrect to rely on IA Geurt Gijssen for answers to questions concerning USCF rules. While IA Gijssen’s credentials as a FIDE arbiter are impeccable, IA Gijssen answers questions directed to his column based on the FIDE Laws of Chess. Only recently has he started to refer to the Official Rules of Chess, and then (I think) only to the rules update document that is available on the USCF web site.

8

This is not specific to this rule (or even to rules about chess), but complex bodies of rules are often full of situations in which several rules apply and may conflict with each other somewhat, which can result in answers that are all a little wrong.

9

One must know all of the circumstances under which the move(s) were made. Generally, I have followed the idea of the “intent” of the player to castle. That is usually easy to determine. However, the little “rules lawyers” at scholastic tournaments like to cause mischief with newbie opponents in order to rattle them. They like to argue for the sake of it and the volunteers can sometimes become confused causing even more problems. Then you have to come in and resolve the matter and state that “intent” is at the heart of the rule. When I deal with such incidents, I inform both players of the correct way to castle, point out the irregularity of touching the rook first but allow castling.

What to do about the players who grab both the rook and the king? This is a habit of many European adult players. If you are not going to follow “intent”, then you are going to have to determine which piece was touched first. Must we install instant replay cameras with slo-mo to determine how many milliseconds difference elapsed between the touching of each piece?

I witnessed such an incident over the touching of the rook first between two elderly gentlemen at a chess club during a casual game. They called a member who usually directed tournaments at the club over for a ruling. The player who touched the rook first told the director he would abide by any ruling. Then he leaned across the board and told his opponent, who BTW was a lawyer, that after the game he would see him outside to finish the dispute. It was not surprising that the lawyer decided to “settle” and withdrew his claim. The director’s decision was unique. He turned over both of their kings and told them that he would be calling their mothers about their sons’ behavior. The whole club erupted in laughter.

10

I would hope the variation stays in place. If one is notating on a paper scoresheet, this should be
a benefit of such. I really see very little incorrect about the idea of instructing a student to a. examine procedures, b. write the move, c. examine procedures once again, and then, make the move
if it follows procedures, else change the move. Writing moves does not by any means, nor should it
dictate what a move should be. Only touching a piece with the intent to move, or touching the opponents piece with the intent to take, does this.

Rob Jones
Senior TD

11

You deliberately teach your players procedures that can get them in big trouble if/when they are ready for events played under FIDE rules??

IMHO this variation should be deprecated in the USCF rulebook.

12

It is deprecated at my tournaments.

I acknowledge the rule is at variance from a (very poor) practice that has been taught for years, which is why I only call a player on it upon a complaint (as we’re supposed to in all but egregious circumstances), and which is why I won’t do anything beyond a warning for the first n violations, where n is somewhere greater than one and less than twenty.

But my leniency only goes so far, and a player does not want to commit the (n+1)th infraction of the rule in one of my events.

13

Maybe we should take a page from soccer and hand out yellow cards.

14

I recall one GM whose scoresheet had multiple crossouts often on the same move.

15

fyi:

15A. Manner of keeping score.
In the course of play each player is required to record the game (both the player’s and the opponent’s moves), move after move, as clearly and legibly as possible, on the scoresheet prescribed for the competition. Algebraic notation is standard, but descriptive or computer notation is permitted. The player must first make the move, and then record it on the scoresheet. The scoresheet shall be visible to the arbiter (tournament directors) and the opponent throughout the game. See also Chapter 3, Chess Notation; 13C3, Filling in moves with flag down; 13I, Refusal to obey rules; 35F6, Scorekeeping options; and 43, Scoresheets.See also Chapter 3, Chess Notation; 13C3, Filling in moves with flag down; 13I, Refusal to obey rules; 35F6, Scorekeeping options; and 43, Scoresheets.

TD TIP: While the rule’s wording indicates making the move first and keeping the scoresheet (paper or electronic) visible at all times, it brings the USCF in alignment with FIDE procedures and sooths many of the fears surrounding electronic scoresheets—see new Rule 43— it is a huge change for many players. TDs are advised to first (and possibly second and third) issue warnings to players that do not comply with this revised rule before enforcing any time penalties (1C2a. prescribes adding two minutes to the opponent’s unused time).

15A. (Variation I) Paper scoresheet variation.

The player using a paper scoresheet may first make the move, and then write it on the scoresheet, or vice versa. This variation does not need to be advertised in advance. The scoresheet shall be visible to the arbiter (tournament directors) and the opponent throughout the game.

TD TIP: TDs may penalize a player that is in violation of 20C, Use of notes prohibited20C. “Use of notes prohibited” if the player is first writing the move and repeatedly altering that move on their scoresheet before completing a move on the board.

TD TIP: When TDs enforce the “visible scoresheet” portion of this rule they are advised to first warn players that do not make their scoresheet (or score keeping device) visible to the TD and their opponent.

16

Coaches still must be training their pupils to cover their moves while writing them down and close their score books. Maybe some training of those Certified Coaches is in order.

17

Suppose the square g1 is attacked when White tries to castle? If the King is touched, a King move must be made, if Rf1 is played the K is still open to castling on the other side. (Perhaps a position out of a King’s Gambit, where Black plays an early …Bc5.)

Not saying its likely - but this would seem to fit the bill.

18

Well, it seems I have my answer. It is acceptable not merely to touch, but to actually move, the rook first. As for determination of the move, it’s a bit fuzzy, but unlikely to ever cause a problem. The most common case where a TD would actually have to make a ruling would be a claimant insisting that someone cannot castle because he moved his rook first, and the answer is that that is irrelevant.

(And I pity the poor TD who finds a player who moves Rf1, lets go of the rook, and then grips the king, but doesn’t castle with it, then writes down Rf1 and hits the clock.)

19

Once upon a time in Florida I was directing a small (by today’s standards) CCA tournament where one of the games ended in O-O mate. The white king was on f6, and Rf8 alone would not have been mate because then the white king could have gone to g7.

I saw that possibility coming about five moves ahead of time, and thought it would be a good idea for me to discreetly check the scoresheets over the players’ shoulders, while the game was still in progress, to make sure black had never moved his king nor rook.

Sure enough, upon O-O mate, white exclaimed, “Huh? That’s still legal?” It was, and eventually everybody had a good laugh.

Back to the main topic, whether a player may castle by moving the rook first, FIDE is stricter than USCF in this regard. And let’s face it, the FIDE way ends some of the potential hypothetical disputes that have come up in this thread. OTOH, it seems a little silly to force a player to move just Rf1 when his intention to castle is obvious, such as when the king is touched immediately after the rook has been set down on f1.

But the common sense approach has its pitfalls, too.

All I know is, I’d hate to see a FIDE arbiter force a player to move just his rook, even if neither player wants that ruling. But it could happen.

Bill Smythe

20

Mike, I do believe many of the rule variations we have in USCF make our set of rules far superior
to that of FIDE. In several cases, FIDE is simply unreasonably inflexible. It is my hope that fewer
and fewer tournament directors in our federation will have anything to do with FIDE as the unreasonable insistence on strict adherence to FIDE rules on the part of the international body
increases. If FIDE refuses to accept dual rated events run under USCF rules, then the answer of
more and more tds I hope is quite clear-- GOODBYE, and don’t let the door hit your ________ on
the way out. USCF can live very well without FIDE arrogance.

Further what is important is the development of the skills of the player–how they analyze and look
at the game. And last, Mike, due to its own inflexibility, it appears that more and more FIDE just may be going the way of the dodo bird and the dinosaur. The only question is just how long progressive
western nations are willing to put up with FIDE nonsense.

Rob Jones