Die Schwalbe

2 problem(s) found in 3828 milliseconds (displaying 2 problem(s)). [COMMENTDATE>=20200919 AND NOT K='Hilfsrückzüger' AND A='Keeble, John Frederick']

1 - P0003417
John Frederick Keeble
2206 The Problemist FCS 16, p. 173, 02/1936
(4+5) C+
Weiß zieht an
1. ... exd6ep 2. 0-0-0 dxe7+ 3. Tf8 exf8=T,D#
play all play one stop play next play all
H.Juel: If Black may castle, last move was not made by Ke8 or Ta8, but by Pd5, so it was d7-d5, and hence White may capture en passant. 1... exd6ep 2.0-0-0 dxe7 3.Tf8 exf8=DT# where the castling (belatedly) justifies the e.p. capture.
Not 1... exd6ep 2.Kd8? dxe7 3.Kc8 e8=DT
C+ by Popeye 4.61, except for the promotion mate dual, which was considered acceptable in the old days.
Nachdruck in "Die Schwalbe" 33 mit Diagrammfehler (wBe6 fehlt).
VL: The first AP-problem! Thematic illegal try:
1. ... exd6ep 2.Kd8? dxe7+ 3.Kc8 e8=T(D)#?? (2007-02-11)
Mario Richter: 46 years between this Problem and P1366663 - perhaps this was not Keeble's first AP-Problem?! (2019-08-29)
Ladislav Packa: Of course I know the idea a posteriori. But let my heresy be forgiven - I do not consider it logical. In this particular example, I consider e.p. for proof that black can castling - not vice versa. (2019-08-29)
A.Buchanan: Hi Ladislav. If ep is ok, then Black just moved a pawn with double hop. That doesn't stop Black having e.g. moved bK right at the beginning of the game. So you can't prove that Black definitely can castle this way. What am I missing here? (2019-08-29)
Ladislav Packa: Your statement is perfectly fine. However, the AP condition is usually interpreted as castling is evidence of e.p. However, the second party does not have to do castling - the possibility has already been demonstrated by the existence of e.p. (2019-08-29)
A.Buchanan: Hi again Ladislav. Please try again, I am obviously being very stupid. I don't understand how 1. exd6ep is any more indicative that Black can castle than 1. K~. (2019-08-30)
Ladislav Packa: Hi, Andre!
e.p. is just a result of the d7-d5 move. Black can castle only if it was the last move of black (even without wPe5). Therefore, it does not have the logic of joining the e.p. with castling. For example, ask yourself if black is allowed to castle in the position without wPe5. (2019-08-30)
A.Buchanan: Hi Ladislav, I still don't understand you. Heresies must be explained extraordinarily carefully :-)
Following the conventions:
(1) Black retains castling rights because we cannot prove that bK or bR ever moved.
(2) If e.p. is now played, then we still require the same castling convention, as bK or bR might have moved earlier in the game.
(3) So the actual execution of the e.p. does not prove anything new about castling.
(4) Hence it is incorrect to say "e.p. [is] proof that black can [castle]".
Which step do you disagree with?
Yes by the convention, Black could castle in the position without wPe5. What does that prove? (2020-10-03)
more ...
Keywords: a posteriori (AP), En passant as key, Castling (sg), Promotion (T), Valladao Task
Genre: h#, Retro
Computer test: Popeye 4.61
FEN: r3k3/2p1p3/2P1P3/2KpP3/8/8/8/8
Reprints: V. Die Schwalbe 33, p. 323, 06/1975
2001 CHM Themes 12/2000
3 Die Schwalbe 215, p. 240, 10/2005
A1 Problemkiste 169 02/2007
14 Die Schwalbe 241, p. 374, 02/2010
Input: Gerd Wilts, 1995-06-03
Last update: Alfred Pfeiffer, 2018-01-03 more...
2 - P1019210
John Frederick Keeble
La Strategie 1900-1902
ehrende Erwähnung
(9+6) C+
1. Tc6! droht 2. Dg5+ e5 3. Dd2+ Sfxd2#
1. ... Sfe3 2. Dd1+ Sxd1 3. Sc3+ Sxc3#
1. ... Sg3 2. De4+ Sxe4 3. Sc3+ Sxc3#
1. ... e5 2. Dg2+ e4 3. Dd2+ Sfxd2#
play all play one stop play next play all
SCHRECKE: C+, popeye 4.85
1.Tc6! (2.Dg5+ e5 3.Dd2+ Sf:d2#)
1. ... Sfe3 2.Dd1+ S:d1 3.Sc3+ S:c3#
1. ... Sg3 2.De4+ S:e4 3.Sc3+ S:c3#
1. ... e5 2.Dg2+ e4 3.Dd2+ Sf:d2# (2020-11-03)
Keywords: Brian Stephenson Collection (4048)
Genre: s#
FEN: 1B6/4p3/4R3/1P1k4/N1n3QR/8/BpN4p/1K3n2
Reprints: Les Mille et un Mats Inverses 1907
Input: Brian Stephenson, 2004-08-12
Last update: Marcin Banaszek, 2020-11-07 more...
Show statistic for complete result.

The problems of this query have been registered by the following contributors:

Gerd Wilts (1)
Brian Stephenson (1)