A Former Loophole-The “Legal” Triple Check

[Rewan Demontay]

The old versions of the FIDE rules of chess said that, regarding check, “The king is in check when the square it occupies is attacked by one or two of the opponent’s pieces.” This left an a potentional ambiguity-if the king was left attacked by three pieces then it was not in technically in check-it was is left in a “legal” triple check.

The change in the wording goes back to the 1984 FIDE Chess Congress since the wording "one or two" first appeared in the 1985 laws of chess http://aekphotography.co.uk/CAA/1985_La ... _Final.pdf This was the wording for the remainder of the 80’s, and it was finally reworded at the 1992 FIDE Chess Congress, as indicated by new phrasing" one or more" in the 1993 FIDE laws: http://aekphotography.co.uk/CAA/FIDE_Laws_1993.docx

Beforehand, the 1980 laws-http://aekphotography.co.uk/CAA/Laws_of ... pdf-merely stated "The king is in check when the square it occupies is attacked by an enemy piece." Even this could considered an ambiguity if one literally interpreted "an" meaning exclusively one piece therefore making it "legal" to leave your king in a double check, although this arguement is reasonably dismissive. It's wonder why they later changed it to "one or two," but that still was ambiguous enough, forcing them to later change it still!

These old versions of FIDE laws, come from this website: http://www.chessarbitersassociation.co.uk/index.htm

Lately, I’ve been tracking down the history of the idea. It seems that Kevin Thurlow, here with us in the forum, was one of the first to notice, and he composed a mate in 1 example that was published in Richard James’s and the late Michal Fox’s CHESS Magazine colum “Addict’s Corner.” Which issue is forgotten to him along with his original version it seems, after I PMed him, but l
have luckily discovered both of those for him. The given position corresponds with what he generally remembers the scheme being.

In an old forum thing archived by Google, I found an exact mention, which I will quote here: https://groups.google.com/forum/m/#!top ... CP3voSjXkw

“CHECK THIS OUT!
"Right, N. Short, J. Speelman, J. Nunn, and the rest of you clever clogs. Break your heart on this: White: Ke3, Qb2, Ne5 Black: Kh8, Re8, Rg8, Ba7, Nb6, Pg7, h7 Black incautiously played a double check 1...Nc4+?? How did White crash through to victory? Admit it; you're baffled. Deduct one hundred Elo points [and read on]:

Dead easy. White gets out of check by 2 Nf7 mate. Yes, it's perfectly legit. Consult Article 9.1 of the laws: "The King is in check when the square it occupies is attacked by one or two of the opponent's pieces..." It says nothing about a threefold attack, so White's move takes him out of check..." -Mike Fox and Richard James in their Addicts' Corner column in the February, 1992 issue of "Maxwell Macmillan Chess (this is one of th several rebrands that the magazine went through).”

Here’s a helpful diagram.

4r1rk/b5pp/1n6/4N3/8/4K3/1Q6/8 [pos]

[Rewan Demontay]

(continued from above)

I’m sure that it would be nice to see the original documentation of course.

Kevin Thurlow did a nice job, along with Stewart Reuben (another member here), for noticing and correcting this neat nugget in the FIDE laws. Kevin's problem reignited the world's notice of the ambiguity, akin to the effect that Tim Krabbe's joke castling problem had, which is what finally promoted FIDE to change the wording of the laws on check.

However, it seems that Kevin was not the first. I thereby present a problem that from 1988(!) that may surprise you Kevin!

Robert Norman, Pergamon CHESS (again, an ealier branding of CHESS Magazine), June 1988
White To Move And Win
1n5k/1r6/5K2/4Q1Pb/8/8/8/8 w - - 0 1

1. g6 Nd7+ 2. Kf7+ Nxe5++ 3. g7+! White exposes themself to a triple check, therefore not technically being in check at all! 3... Kh7 4. g8=Q+ Kh6 5. Qg7#

I have found it in only a few place online, with the most detailed being a 4/1989 editon of the EG magazine on page 24: http://www.arves.org/arves/images/PDF/EG_PDF/eg95.pdf
On A Dutch Site: http://www.caissa-amsterdam.nl/caissa-p ... n/#respond
A Peter Norman Obituary: https://www.keverelchess.com/exmouth/pioneers/3199/

There exists a very beautiful 2008 variation of the Norman problem created by IM Jens Kristiansen, from vague memory it appears, on their old blog post about various jokes based on rule exploits over the course of chess’s exsistence, which is what started me on this quest: http://faroffchess.blogspot.com/2008/12 ... rules.html

White To Move And Win
7k/rn6/5KP1/p1p4b/8/8/8/7R w - - 0 1

**********
This about all I know of the old triple check possbility from what research I have done. I hope that you enjoyed learning a bit about chess history! Any more information about the origins and misty history regarding this subject would be greatly appreciated, no matter how small! Feel to leave a reply of any kind.

These are also my first posts on this site.

[John McKenna]

Thanks, I sometimes wondered if double check was the most that could be delivered legally and did not consider illegal moves. Discussions (between arbuters and players) of which are to be found on this forum.

There's a typo in the moves given above to the Robert Norman diag.

Not 1.e6 Nd7+...

But 1.g6...

[Rewan Demontay]

Many thanks, and typo fixed.

[Matt Mackenzie]

There have been problems inspired by another previous "loophole" of the Laws - they never originally specified that you had to promote a pawn to a piece of the *same* colour. Of course, the spoilsports at FIDE have now fixed that one as well.

[David Sedgwick]

(continued from above)

I’m sure that it would be nice to see the original documentation of course.

Kevin Thurlow did a nice job, along with Stewart Reuben (another member here), for noticing and correcting this neat nugget in the FIDE laws. However, it seems that Kevin was not the first. I thereby present a problem that from 1988(!) that may surprise you Kevin!

Robert Norman, Pergamon CHESS (again, an ealier branding of CHESS Magazine), June 1988
White To Move And Win
1n5k/1r6/5K2/4Q1Pb/8/8/8/8 w - - 0 1

1. g6 Nd7+ 2. Kf7+ Nxe5++ 3. g7+! White exposes themself to a triple check, therefore not technically being in check at all! 3... Kh7 4. g8=Q+ Kh6 5. Qg7#

I have found it in only a few place online, with the most detailed being a 4/1989 editon of the EG magazine on page 24: http://www.arves.org/arves/images/PDF/EG_PDF/eg95.pdf
On A Dutch Site: http://www.caissa-amsterdam.nl/caissa-p ... n/#respond
A Peter Norman Obituary: https://www.keverelchess.com/exmouth/pioneers/3199/

There exists a very beautiful 2008 variation of the Norman problem created by IM Jens Kristiansen, from vague memory it appears, on their old blog post about various jokes based on rule exploits over the course of chess’s exsistence, which is what started me on this quest: http://faroffchess.blogspot.com/2008/12 ... rules.html

White To Move And Win
7k/rn6/5KP1/p1p4b/8/8/8/7R w - - 0 1

**********
This about all I know of the old triple check possbility from what research I have done. I hope that you enjoyed learning a bit about chess history! Any more information about the origins and misty history regarding this subject would be greatly appreciated, no matter how small! Feel to leave a reply of any kind.

These are also my first posts on this site.

I first saw the 1988 version when it was demonstrated by IA Paul Lamford at a quarterly meeting of the Chess Endgame Study Circle in London. I think that that must have been in June 1988 or September 1988.

In October 1988 I showed the study to Sofia Polgar and Judit Polgar at the Duncan Lawrie World Mixed Challenge, also in London. They were not impressed.

[Richard Thursby]

Of course, if a king is attacked by three of the opponent's pieces then it is definitely attacked by "one or two" of them (as in, there exist one or two opponent's pieces attacking it).

[David Sedgwick]

Of course, if a king is attacked by three of the opponent's pieces then it is definitely attacked by "one or two" of them (as in, there exist one or two opponent's pieces attacking it).

If I say that a Queen is attacking two pawns, does that mean that it could be attacking three pawns?

[Richard Thursby]

If I say that a Queen is attacking two pawns, does that [mean] that it could be attacking three pawns?

Correct. It is like the following: Some months have 30 days, some have 31 days. How many have 28 days?

[Rewan Demontay]

If you notice I wrote specifically “ This left an a **potentional** ambiguity,” notice the added asterisks, to cover such logic.

The reprint in the EG magazine of the 1988 problem also made note of such reasoning: “ Well, one can easily counter that check by three men includes check by two and is therefore covered (not 'specifically excluded') under the Laws, which do not need to be changed.”

But nethertheless, we can surely all agree that this unintended interpration of the rules at the time resulted in an amsuing and hilarious potentional “exploit”?

[David Sedgwick]

It is like the following: Some months have 30 days, some have 31 days. How many have 28 days?

One, or zero in a leap year.

[Rewan Demontay]

It is like the following: Some months have 30 days, some have 31 days. How many have 28 days?

One, or zero in a leap year.

Not quite David-all months have 28 days. The point/analogy Richard is making is that all 3s have 2 in them, so henceforth does the notion, such as Feburary being the only month with 28 days, of a legal triple check really exist? (I don’t necesssarily agree since it is a lovely gem that deserves to exist, but I understand it enough to be able explain it. All months definitely have 28 days though.)

The idea of a legal triple trick was witty nitpick on chess law that was cleverly used to create diamond joke chess problems Regardless of you believe if the triple check was legal or not at the time, the ambiguity was strong enough for FIDE to change the wording to “one or more,” which definitely says something and gives it humorous value.

[Kevin Thurlow]

My feeling was that the Laws should be unambiguous and say what they mean. If "check" is defined as attacking the king with "one or two" pieces, that definitely excludes check from more than two pieces.

I also managed to get the Laws of Hockey changed and rather more importantly, the regulations relating to the international transport of dangerous goods.

[David Sedgwick]

My feeling was that the Laws should be unambiguous and say what they mean. If "check" is defined as attacking the king with "one or two" pieces, that definitely excludes check from more than two pieces.

I also managed to get the Laws of Hockey changed and rather more importantly, the regulations relating to the international transport of dangerous goods.

I agree with Kevin. I disagree with Rewan and with Richard.

It is about 60 years since I first learnt that there is only one month which has 28 days.

If someone asked me how many months have 27 days, I would reply "None" rather than "Twelve".

[JustinHorton]

Yeah, but it's possible that the months question is a poor comparison. We know all the months have at least 28 days, and that when one asks "which month has 28 days" they are either seeking the answer "February" or asking a trick question. We also know that nobody intended to frame the Laws of Chess such that we could escape check by moving into three simultaneous checks, because it would be ludicrously silly.