Julia's Fairies

No.627 (SL)

Sébastien Luce 


Original Problems, Julia’s Fairies – 2014 (III): September – December

→Previous→Next ; →List 2014(III)

Please send your original fairy problems to: julia@juliasfairies.com

No.627 by Sébastien Luce – Valladao theme in Symmetry Circe! (JV)


Symmetry Circe: As Circe except that the rebirth square for the captured unit is that which lies at an equal distance (in a straight line) beyond the midpoint of the board. Thus a capture on c4 produces a rebirth on f5, a capture on g1 produces a rebirth on b8, and so on. Strictly speaking there are of course other types of symmetry: this one is rotational.

Circe: Captured units (not Ks) reappear on their game-array squares, of the same colour in the case of pieces, on the file of capture in the case of pawns, and on the promotion square of the file of capture in the case of fairy pieces. If the rebirth square is occupied the capture is normal.

No.627 Sébastien Luce

original – 18.10.2014

Solutions: (click to show/hide)

white Kf1 pg2h3 black Ke8 Ra8 Sd8 Pa6b6h4

h#4,5                                    (3+6)
Symmetry Circe

Notify of

Newest Most Voted
Inline Feedbacks
View all comments
October 19, 2014 22:27

Beautiful! The composer has skillfully added an excelsior also, making it an unique task!

Dominique Forlot
Dominique Forlot
October 19, 2014 23:13

Wow! just Crystal clear !
All white moves by this only magical pawn !!
This problem would deserve a place in an album FIDE.

Geoff Foster
Geoff Foster
October 20, 2014 01:11

An interesting way to force the double-step of the pawn! However, the capture occurs on g3, so to me the rebirth should be on b6, not b5. The effect of an en passant capture should be the same as the effect of a normal capture.

October 21, 2014 21:40
Reply to  Geoff Foster

Of course the capture is on g3. What is relevant for reappearance is where the piece was when it was captured. If a locust captured the pawn on g4 and landed on g3, I am sure this doubt will not arise.

Sebastien Luce
Sebastien Luce
October 20, 2014 01:46

Thank you to all your comments !
It is strange because I created this compo very quickly and
I proposed other problems to Julia to publish !

Geoff, precisely this problem uses the difference between 1…g3 and 1…g4. This stratagem is specific in Symmetry Circe. Very often, composers use the addition of Madrasi to “force” the double advance of the pawn but it is an easier way.

Dominique Forlot
Dominique Forlot
October 20, 2014 11:07

In this type of circe the pawn has to be reborn in reference to the square “reached during the capture”, so to me the rebirth should be on b5.

the symmetryCirce is a wonderful Idea for this task : the pawn’s rebirth on b5 and the (virtual) Queen’s rebirth on g2 are very exquisite.
a cute”Thrown effect” on the chessboard.
I like this problem.

Dominique Forlot
Dominique Forlot
October 20, 2014 12:11

I find a Baby for the O-O !

FEN: 4k2r/K7/8/8/6p1/8/5P2/8
white Pf2 Ka7
black Pg4 Ke8 Rh8

h#4,5 (2+3)

1…f2-f4 2.g4*f3 ep.[+wuPf4] uPf4-f5 3.0-0 uPf5-f6 4.Kg8-h8 uPf6-f7 5.Rf8-g8 uPf7*g8=uQ #

Valladao theme With o-o ( not o-o-o ) & excelsior too.

Sebastien Luce
Sebastien Luce
October 20, 2014 22:06

c’est effectivement un tres joli bébé !
(néanmoins si j’ai bien compris en “fantomes” quand une piece a capturé, elle ne peut plus être capturée ce qui donne lui donne un pouvoir un peu excessif a mon goût)

Dominique Forlot
Dominique Forlot
October 21, 2014 00:15

oui, je dois reconnaître que la puissance de la pièce fantôme est redoutable( trop sûrement ) mais elle permet ici d’utiliser le matériel minimal pour le thème ( KPkrp) juste pour le fun donc.

La force de la condition Circé est également fantastique car elle permet des « pas » considérables.
Dans ton problème par exemple, le pion gagne un tempo pour son Excelsior.

Avec la condition “circeAntipodes” nous pouvons par exemple pousser le trait et d’obtenir l’Excelsior en deux coups de pion seulement !

Antipodean antiCirce: As antiCirce but the rebirth square for the capturing piece is the one at a distance of 4,4 from the square where a unit is captured (its “antipodes”).
For c5 the antipodes is g1, for e2 it is a6 and so on. This square must be vacant, else the capture is illegal. Pawns reborn on promotion squares promote immediately.


White : Kb3 Pa2
Black : Ke8 Qe6 Ra8 Pd7d5b4

h‡3 (2+6)

1.0-0-0 a2-a4 2.b4*a3 ep.[+wPe8=Q] Qe8-f8 3.Rd8*f8[+wQb4] Qb4-b8 #

toujours dans un esprit d’échange et pour explorer encore ce thème Valladao.
Amicalement 😉

October 21, 2014 00:46

Like Geoff, I wonder which is the underlying logic allowing, after g2-g4, this pawn to be en-passant captured on g4, rather than on the more customary square g3.

Indeed, in e.g. T&M, the en-passant capture following g2-g4 is h4xg3->g4, and not h4xg4->g5. Hence the capture occurs on g3, not on g4.

What is the reason for not keeping such a classic rule in symmetry circe?

Sebastien Luce
Sebastien Luce
October 21, 2014 10:35

Dear Dominique,
if you are interested by the Valladao theme, there is a tourney organised by Best Problems actually (in series movers).
It is the occasion to publish Originals.

Sebastien Luce
Sebastien Luce
October 21, 2014 10:49

To answer to the comment of M.Dupont and Geoff
with Black: h4 White: g2
after g4 hxg3 (e.p.) if the capture is indeed in g3, the captured pawn is already in g4 !
This light nuance explain the “effect” in Symetry Circe.

October 21, 2014 18:50

I find this symmetry circe rule concerning e.p. a bit confusing and artificial, but never mind, this is just a matter of taste.

A more important point is that the definition should mention this exception, otherwise the reader who is not aware of it has no chance to solve the problem!

Sebastien Luce
Sebastien Luce
October 21, 2014 20:05

Dear M. Dupont, it is not artificial, it is logical !
with Black: h4 White: g2 after 1.g4 hxg3 (e.p.)

A) in T&M the capturing black pawn h4 has to play a white pawn move after capture on g3 and go in g4 !
B) in Symetry Circe the captured white pawn g4 has to rebirth to the symetrical square b5 !

I hope everything is clear.

Dominique Forlot
Dominique Forlot
October 21, 2014 20:32

As Sébastien Luce, I used the same peculiarity In my two problems above and I think It’s justified to use it in this way.

[ GhostChess October 20, 2014 at 12:11 ] & [AntipodeanCirce October 21, 2014 at 00:15 ]
they are C+ by WinChloe 3.29 & Popeye 4.69

the same rule is applied in other fairies conditions and I can find a strong logic for this.

the “prise en passant” is an convention to accelerate the play of pawns!
if white choose 1.g2-g4 to build an idea his pawn is on g4 ( not on g3).

now if Black have a pawn on f4 (h4) he is not hurt because he can again capture the pawn by F4xg3! (h4xg3!)… Indeed the capture is on g3! the acceleration of the white playing don’t have to be a deceleration of the black playing.

But to remove the pawn of the chessboard it is necessary to take it in square g4… because of the White choice to play g4!? not g3?!

Why the composer should him forget the choice made in the previous move and and consider that g4 was never played?

The rebirth of the pawn in the fourth row is also justifiable as its rebirth in the third row… but the first hypothesis keeps the memory of the move really played.

“…and nevertheless g4 was played!!” Galileo Galilei 🙂

Sebastien Luce
Sebastien Luce
October 21, 2014 22:37

To be completely clear about “en passant capture” :
if you are playing a chess game with black and you execute an “en passant capture” hxg3 (e.p), what do you do ?
You put your black pawn on g3 square and then after you take of the white g4 pawn !

October 21, 2014 23:10

Ok. But I still think that, for the convenience of the solver, it should be noticed in the definition that the departure is not the square where the piece is captured, but the square reached by the captured piece.

Btw, do fairy pieces exist with “ep power”? (except the orthodox pawn!). For example an ep-Bishop in d5 would be able to capture on f3 a Queen playing Qd1-h5. This question is for you Dominique, the fan of new notions! Would be interesting to construct examples which motivate the use of ep power…

Dominique Forlot
Dominique Forlot
October 22, 2014 00:48

“ep power”?
Oh yes, beautiful new idea Nicolas! 🙂

I think the king (orthodox) is a kind of “ep power” piece too! Because it can’t castle if the squares to be crossed are under control.(squares where the king should be captured “en-passant”… Unbearable for king)

This characteristic could be extended to all the pieces besides the pawn and the king!
“Capture in the flight or secured after landing”

Surprising and interesting extrapolation by your prolific brain (^.^)

October 22, 2014 01:33

Thanks Dominique. I’m gonna think a little bit more about “ep-pieces”, and then probably open a thread on FE… You might begin, by your side, to search for interesting applications, in particular how to motivate that a given unit is forced to be ep-captured after a long move, rather than just playing normally on its capturing square… To undo a check or to capture something are the trivial motivations, but it should exist some deeper ones.

Nikola Predrag
Nikola Predrag
October 22, 2014 22:37

“…Dear M. Dupont, it is not artificial, it is logical!
with Black: h4 White: g2 after 1.g4 hxg3 (e.p.)…”

Well, it doesn’t look logical. Pawns can’t capture orthogonally on the same rank, that would be quite a fairy interpretation of e.p. capture. And finishing the move on a square different than a square of capture, would multiply the fairy interpretation.
I don’t believe that en-passant was invented with such reasoning.

The exceptional ability of a wP/bP on 2nd/7th rank is compensated by the exceptional ability on 5th/4th rank.

The Pawn is a piece allowed to move only one step. Double-step is possible in two moves.
One-move double-step g2-g4 from the initial rank can’t be treated as a “monolithic” move, because bPh4 would not be able to capture wPg4.
The way of writing the move, hxg3ep., shows a legal capture of wPg3 by bPh4.
This could be interpreted as: wPg2 makes a move compound of two parts, one “foot” steps” on g3 and while it’s still there, the other “foot” steps on g4.
bPh4 is able to see the rear “foot” and catch it before it is pulled to g4. Then the whole wP is dragged to g3 and annihilated. That happens before the white move is completed, wP actually never reaches g4. The head is above g4 but the chest is not, as in the athletics. The weight of the body does not leave the square g3.
The other pieces are not affected by (nor aware of) that “one-foot-step” on g3, they see wPg2 before the move started and wPg4 after the move is completed. Their “cameras” are to slow to record the details of “photo-finish”.

bPh4xwPg4-g3 might be legal if bP is defined as a fairy piece or if such capture is defined as a “fairy e.p.”.

Kostas Prentos
Kostas Prentos
October 24, 2014 03:55

In orthodox chess, the en passant capture can be described in simple words as: “The pawn that played a double step is captured as if it had played a single step”. As a result of this definition, the black pawn h4 ends up on the square g3 and the white pawn disappears (from the square g4, where it was last standing).

In fairy chess, it is not so simple. The white pawn moves to g4 and the black pawn captures it on g3. I would analyze this move sequence in the following way: The white pawn was captured as if it moved one step (on g3), but because it occupied g4, this square should be considered as the square from which it disappeared. The capturing pawn ends up on g3, because on that square the capture took place. Using this analysis, it is easy to explain why in Take & Make the correct move sequence is 1.g2-g4 h4xg3-g4 (not g5). The reason is that the capturing black pawn plays a white pawn move after the capture on g3, from the square on which the capture took place (g3). The fact that the captured pawn stood on g4 does not matter for Take & Make.

In different fairy conditions like Symmetry Circe or Circe Parrain, it actually matters where the captured pawn was when the e.p capture took place. So, in the case of Circe Parrain, the following sequence 1.g2-g4 h4xg3 e.p. 2.a2-a4(+wPg6) makes sense, because the white pawn was captured when it was standing on g4, even if the capture occurred on g3. I do not see why it should be different in Symmetry Circe. The rebirth square of a unit that was captured when it was standing on g4, should be on the symmetrical b5 and not on b6 (just because the capture occurred on g3). In my view, like that of the composer and the main testing programs, this problem is sound. However, it is not an easy call. I understand the reservations that many other problemists expressed. Unless clear and detailed definitions for each fairy condition and all the special cases (castling, en passant capture, promotions, pawns on the first rank, etc.) are universally accepted and become official by a central authority, we can argue and improvise until Kingdom come.

Nikola Predrag
Nikola Predrag
October 24, 2014 10:36

bPh4 simply does not attack square g4 and can’t capture any piece on g4!
But bPh4 attacks square g3 and can capture a wP while it is PASSING ACROSS g3, BEFORE reaching g4.
This is rather simple.
After g2-g4, wP only appears as standing on g4, but it is not actually there until bPh4 accepts and confirms that as a legal completed move.
bPh4 has the exceptional right to announce a capture hxg3ep. then the move by wPg2 is completed as g2-(g4)-g3 and Black is obliged to execute the announced capture.

This is by far simpler than saying: bPh4 captures wPg4 and arrives on g3.
I believe that the only logical approach is that wP is indeed captured on g3, while passing across it and before arriving to g4.

In case of Madrasi, the “photo-finish tale” from my previous post might explain that wP is not on g3 with “both feet” and therefore it has not the power as it would have if standing on g3 with “two feet”.

Circe Parrain’s interpretation is not logical too.

Georgy Evseev
Georgy Evseev
October 24, 2014 14:24
Reply to  Nikola Predrag

I agree with Kostas here.

There is no question that pawn arrives to g4, as it can defend against, for example, bishop check, and such a move is definitely legal.
So, white pawn definitely disappears from g4, and as a result its correct rebirth square is b5.

There is a beautiful Circe Parrain problem by Michel Caillaud which should be in AF01-03. It also uses very non-standard approach to e.p. capture and, being the judge then , I have given to it 4 points (which means that I wanted to see this problem in an Album, no matter what the other judges thought).

I do not have the book at the moment, and so unable to give the exact number or to cite the composition. The interpretation of rules in it also may be questioned, but the brilliance of idea justifies this interpretation from my point of view.

October 24, 2014 14:31
Reply to  Georgy Evseev

The e.p capture is indeed a fairy capture. It is only occasion in orthodox chess, where the capture does not take place on the square where the victim is. That is why I gave the example of capture by Locust !

Nikola Predrag
Nikola Predrag
October 24, 2014 17:07

Yes, parrying the check, as Georgy has mentioned, might prove that at least one part of wP (sufficient for shielding) arrives to g4.
But a part of wP which is crucial for the capture is still on g3.

What “en passant” means? The change in the rules should be expected to retain the main logic of the previous rules, before the ep. capture was introduced.
wPg2 must PASS across g3 to reach g4, that’s an obvious and consistent logic. If g3 is occupied, wPg2 can’t move to g4.
hxg3, while wP is, at least partially, still on g3, also retains most of the old logic.
h4xg4-g3 might be interpreted as bPh4 passes across g4, annihilates wP, and arrives to g3.
Such logic is not valid for any other captures by a Pawn.
It’s not likely that bPh4 can pass across g4, but wPg2 can quite logically pass across g3.
So, what “en passant” means?

If wP standing half on g3 and half on g4 is not an acceptable explanation, I would like to hear a more acceptable one?
Traditional rules treat a capture as it happens on the arrival square of the capturing piece.

Even the fantastic “time-traveling backwards” would preserve the logic consistency more than h4xg4-g3.
g2-g4 legally parries a check and bPh4 captures it on g3, changing the past. The result in “present time” is also a check with White on the move.
wP has dissapeared from g4 but as a result of the capture on g3.
Is the rebirth square related to a square of capture or to a square of disappearance?

Amusing speculations 🙂

October 24, 2014 19:57

In fact the precise meaning of the sequence g2-g4 h4xg3 is a matter of convention, depending on which move is considered “fairy”. Roughly speaking, there are 2 different possible ways:

A: g2-g4 is orthodox (hence the white g-Pawn actually reached square g4), and the capture h4xg3 is fairy, as it occurs on the square g3 where no piece is standing.

B: g2-g4 is fairy, and might be denoted g2-(g3 or g4), as we have to wait for the next black move to decide which square is reached. If this black move is not an ep-capture, then g2-g4 was earlier played. But if this black move is an ep-capture h4xg3, then g2-g3 was earlier played (hence the capture is orthodox and the white g-Pawn actually never reached square g4).

I slightly prefer convention B (a friend of mine called g2-g4 a “quantic” move, as we don’t know yet if this pawn is standing on g3 or on g4), but I can easily live with convention A as, anyway, some fairy trick has to be used in both possibilities.

Note that, in the situation where a check must be parried on g4, convention B is still consistent. Indeed the move g2-g3 is enough for the job, as it is followed by h4xg3, and hence black can’t at the same time capture the white King!

Georgy Evseev
Georgy Evseev
October 25, 2014 16:47

The problem I mentioned is G194 in AF 01-03.

Kevin Begley & Michel Caillaud
9 H.M., StrateGems 2001
White : Kg1 Rh3 Bf3 Ph4 (4)
Black : Ka6 Bb7 Sa7b5 Pf4g4e3g3g2 Gh1 (10)
Circe Parrain
Grasshopper (0+1)

1.Gxh4 Rh1(h2) 2.Gxh1 h4(Rh3) 3.gxh3 (with e.p.!) Bxb7(d8G, Rd7)#

Nikola Predrag
Nikola Predrag
October 25, 2014 18:14
Reply to  Georgy Evseev

The rules have to be unambiguous, so what are the rules here?
Capture of some piece on one square and arrival to an occupied square with additional capture of the occupying piece?
“…A pawn attacking a square crossed by an opponent’s pawn which has advanced two squares in one move from its original square may capture this opponent’s pawn as though the latter had been moved only one square….”

Not a perfectly clear rule, but the consistent interpretation would be that “…the latter is captured as though it had been moved only one square…”

This not only tells where the Pawn was captured but also
that the rebirth of wR would be on h2 (as though the 2…h2-h3 was played!)

So, which rules are hidden behind Circe Parrain? And do you consider the logic of these rules as unambiguously evident to a solver?

Georgy Evseev
Georgy Evseev
October 25, 2014 23:18
Reply to  Nikola Predrag


There is no way the inventor of new fairy condition may ensure there no “hidden rocks”, potentially ambiguous situations or unexpected interactions with other rules.

When a composer finds such “hole”, what he must do? Natural approach is to use this “hole” as an idea for the problem. Then some interpretation of rules appears in composer’s mind, the problem is composed and published.

And suddenly there is no sense in shutting the barn door after the horse has bolted.

Dominique Forlot
Dominique Forlot
October 25, 2014 23:19
Reply to  Georgy Evseev

to obtain in two moves a position visually identical but conceptually different is fantastic
The double capture (R + P e.p.) in g3 is surprising!
this problem is magical for me, with fairies idea in each move!

Nikola Predrag
Nikola Predrag
October 26, 2014 01:35

Georgy, the trouble is with my perception, I don’t see a hole. I see simply a newly invented rule of ep. capture, but not even mentioned in the stipulation, not to speak about the explanation.
I might explain it, for instance:
“In case when the arrival square of a Pawn which makes the ep.-capture is occupied, the pieces which occupy the arrival squares are treated as captured by the arriving pieces.”
And to “compose”:
White Kc3 Pg5 Sf2 Rg1; Black Ke2; Neutral Pf7; Stipulation H#1; Condition Circe AntiCirce
1.nPf5 gxf6ep.(nPf7,wPf6-f2,wSf2-g1,wRg1-a1)#
Rule of ep.capture becomes stronger than the standard Circe+AntiCirce rules.
(Cheylan type wouldn’t allow the “rebirth-captures” on f2 and g1.)

But my perception could improve if the authors’ interpretation would be presented, at least a posteriori.
So, where is the mentioned hole?

Georgy Evseev
Georgy Evseev
October 26, 2014 12:13
Reply to  Nikola Predrag

Author’s interpretation seems obvious to me from the solution. The white pawn is captured normally according to standard ep rules and _at the same time_ white rook is captured normally according to standard rules. It is just happens that this is the same move for black.

The text “as though the latter had been moved only one square” in the rules is used to explain the procedure, not to move the pawn being captured back.

The “hole” is that nobody ever thought before that the “ep” square may have a piece on it. And, as always, everything that was not defined (or imagined) in advance may give birth to a new fairy condition or to a variant of existing one, exactly like AntiCirce Calvet and AntiCirce Cheylan have branched.

Nikola Predrag
Nikola Predrag
October 26, 2014 15:48

“…The white pawn is captured normally according to standard ep rules…”
Well, I am surprized that anyone sees any logic in such an interpretation of “standard rules”.
Absolutelly nothing suggest that bPg4 attacks h4 and that it can capture a piece occupying h4.
According to standard rules, bPg4 attacks and can capture only a piece occupying h3 or f3.
I’m indeed interested to see some serious version of the standard rules, where a Pawn can attack a neighbouring square on the same rank.

And what might suggest that the formulation “as though the latter had been moved only one square” could be interpreted: “as though the latter (wPh4) has been attacked and captured on h4 by bPg4 which is after the capture tranferred as though the capture has happened on h3”?

This doesn’t look as a “fantastic logic” to me but only as an logically unfounded CHANGE of standard rules.
Standard rules define a capture:
“… If a piece moves to a square occupied by an opponent’s piece the latter is captured and removed from the chessboard as part of the same move. A piece is said to attack an opponent’s piece if the piece could make a capture on that square…”

So, a piece can be captured only if it occupies a square to which an opponent’s piece can move.
As long as some EXCEPTION does not explicitly allow bPg4 to move to h4, no piece occupying h4 could be captured by it.

But quite logically, standard definition of ep. capture and general definition of any capture, should be interpreted:
“…A pawn attacking a square crossed by an opponent’s pawn which has advanced two squares in one move from its original square may capture this opponent’s pawn as though the latter OCCUPIES A SQUARE to which it would arrive in case it had been moved only one square….”

Whatever fantastic speculations might be, the standard rules clearly say that the mentioned capture of wP can happen exclusively on h3!
Since no other excepional rules are defined in the standard rules, the “procedure” is subordinated to the defined rules.
Therefore, all consequences of a procedure”as though something has happened” will be”as though something has INDEED happened”.

If the laws treat someone as though he has commited a particular crime, the consequences of a procedure will be as though he has indeed done that, unles something defined in the procedure would allow a different treatment.

If wPh4 would be allowed to claim: “But I don’t really occupy h3”, that would have to be defined in the rules or at least in the procedure.
A possible amendment to the procedure might be “bPg4 is allowed to move to h3, as though it captures wP on that square but wPh4 remains unaffected by that hypothetical capture, since it does not actually occupy h3 and therefore can’t be really captured by bPg4”. But there’s no any amendment.

I still see no hole in the standard rules, the consequences of “may capture…as though… ” will be as the rules define in case of h2-h3, including the rebirth of wR on h2.

I repeat, on what basis anyone would ever think that the square of capture is h4 and that subsequently the rebirth square is b5?
And (what “…nobody ever thought before…”) that “as though g2-g3 was played” can mean “as though both wP and wR are on h3” or “as though wP is on h4 and wR is on h3”?

If that fairy condition is not well defined itself, it still doesn’t affect the standard rules of ep. capture.
If it does, show us the hole in the standard rules.
There are some minor holes, but irrelevant for standard chess as well as for this problem.

Kostas Prentos
Kostas Prentos
October 27, 2014 00:54

I remember the Begley/Caillaud problem from when it was first published. In my view, it is a clear demonstration of the role of a pawn that is captured en passant. The white pawn moves to h4, and at the same time the white Rook is reborn on h3. There is no doubt that the white pawn stands on h4 and that the Rook has to be reborn on h3 according to the Circe Parrain rules of rebirth (unless we prefer the illogical suggestion that both white pieces stand on h3). This completes White’s move: 2…h4(+wRh3).

What happens after this move is irrelevant to which move has already been played and has no effect to it. I mean that Black’s option to capture e.p. does not change anything on the white move that is already complete. Black can either capture the wRh3 only, or capture both white pieces h3 and h4 (e.p. for the Ph4), or play something else. The choice to capture both white pieces h3 and h4 is compatible to the rules of Circe Parrain, has no retroactive effect on White’s previous move(s) and I dare say it should be evident to the solver (or at least, as a possibility), by just following the rules of Circe Parrain.

I can imagine an argument that it is not possible to capture two pieces with one move, but if we ignore this detail for the moment, we can clearly see where these simultaneously captured pieces are going to be reborn: The wPh4 on d8 and the wRh3 on d7 (not both pieces on d7, of course). End of the analysis; now we can argue whether it is possible to capture two pieces with one move, but no matter what one thinks about that, the rebirth square of a pawn that was captured e.p. is related to its former position on h4 and not h3. For the record, I find the logic of the above mentioned problem very clear and I support the composers’ view that it is possible to capture two pieces at once.

Nikola Predrag
Nikola Predrag
October 27, 2014 02:21
Reply to  Kostas Prentos

One thing that is certain in standard rules – bPg4 attacks neither square h4 nor any piece on it!
So, ep. capture can’t take place at h4 and the rebirth can’t be related to h4!
But bPg4 can capture on h3 and that’s where wP is captured.
What happens with wR is another question, but the rules say that ep. capture may be executed “as though a one-step move was played”! And that consequently means “as though wR is reborn on h2”!

All changes of the standard rules have to be explicitly stated in the definition of a fairy condition.
Any other rule remains standard.

You may speculate that
“…Black’s option to capture e.p. does not change anything on the white move that is already complete…”
and wonder “how wPh4 may be captured on h3”.

But if you can’t accept that wP at least hypothetically occupies both h3 and h4 in the same time or “has one foot on each of these squares”, then how on Earth would you come to the conclusion that bPg4 can ever capture on h4???????

Kostas Prentos
Kostas Prentos
October 27, 2014 05:21

Nikola, please read my previous comment one more time. I already explained that the pawn stands on h4 and the Rook on h3 and this cannot change retroactively. The black pawn g4 does not capture on h4, but on h3 (as if the white pawn moved one step forward, but the fact remains that the white pawn played a two-step move and this can’t change). The pawn h4 is captured en passant, not normally.

You can argue that the core idea of the problem is not acceptable by your standards, but not that the pawn and Rook do not stand on h4 and h3, respectively, because they are there-you can see them there. There is no way that any future moves by Black can change this fact (at least not in Circe Parrain) and it is absolutely impossible to shift the white Rook to h2 with a black move.

No matter what we think about the legality of the Begley/Caillaud problem, we can take the opportunity and understand that if the e.p. capture were at all possible, the white pawn’s rebirth would be related to the square h4 and not to h3. There could not be a better example to demonstrate this fact than this particular problem.

Georgy Evseev
Georgy Evseev
October 27, 2014 08:57

My last attempt to explanation.

1. All chess moves are instant. For example when white bishop moves from a1 to h8, it disappears from a1 and appears (out of thin air) on h8. It does not really moves via b2, c3 and so on, as there are no effects which can be tied to moving the bishop over these squares. The same is true for the pawn. The pawn disappears from g2 and appears on g4. It does not in reality NEED to pass g3, even if it is easier to interpret its movement this way.

2. In this case in the en passant capture the capturing pawn arrives to g3, while captured pawn is removed from g4. It is the only case of such non-standard behavior in ordinary chess, and this causes all kinds of difficulties (for new players, for easy explanation in rules, etc.). That is the reason for this discussion also.

3. At the same time there is quite a number of fairy pieces/conditions when the square of arrival of capturing piece differs from the square where the captured piece stands. I have done very fast search for Locust/Circe combination and yes, the rebirth square is defined by the placement of captured piece, not by the square of arrival. For example, bishop captured by Locust frontally keeps its color. So, the solution of 627 is in agreement with general trend.

4. Generally, unless there is something extremely outrageous it is the author’s freedom to treat any fairy condition as he sees correct. While there are traditional fairy conditions, there are no “correct” or “wrong” fairy conditions. When I am acting as a judge, I generally accept everything thrown at me even I do not like specific fairy element (and there is quite a number of elements I do not like))). While I may lower the place for use of elements, their combinations, or interpretations I consider “bad”, I never ignore such compositions and the problem may get high place if there is something in it that overweighs drawbacks.

Nikola Predrag
Nikola Predrag
October 27, 2014 10:46
Reply to  Georgy Evseev

1. Chess was invented and developed on the simple logic extraordinary clear and easily applicable through many centuries.
“…Chess originated in India, where its early form in the 6th century was chaturanga, which translates as “four divisions of the military” – infantry, cavalry, elephants, and chariots, represented respectively by pawn, knight, bishop, and rook….”

What makes you believe that the traditional logic explains the moves as instant?
Only the cavalry can jump over the pieces, all other pieces travel across the squares. That’s why they can’t pass through the pieces.
Pawn is a soldier-pedestrian and can walk only one step at the time.
I say that he stretches one leg to a front square and then joins the second leg to the first one.
From the initial square it is able to stretch the other leg onto the second square, while his first leg is still on the first square. That’s a traditional concept of walking.
He starts to move the first leg towards the second leg on the second square only when the other side has already started to play a move.
Only the opponents Pawn can catch the rear leg in time etc..

Is there more consistent logic in such understanding or the most essential logic, on which the invention of chess is founded, is the “instant moves”?
And the incredible idea that a piece can be captured on a square although the opponent’s piece has never played to this square; what logic would that be?
Such deep change of the most essential logic suggests that it is not chess any more.

The “two-legs” explanation is not important, but the logic clearly allows that a double-stepping Pawn can be on two squares, passing across one and stepping onto the other.
The opponent’s move is also not instant, it begins and before that move is finished, the double-stepping move will be finished.
A Pawn can “catch the rear leg” of the double-stepper” and it falls back onto the rear square and the ep. capture is executed.

Such dynamical, non-instant logic seems convincing to me.
What would make the “instant” logic convincing as a whole?
A set of inconsistent rules with adjusting the logic to explain the separate single features?

Georgy Evseev
Georgy Evseev
October 27, 2014 11:30
Reply to  Nikola Predrag


Historic allusions are good to explain, how the rules have developed, but not how they work now.

Pawn, moving from g2 to g4, does not check black king standing on h4. This is enough confirmation for me that this pawn never really visits g3.

October 27, 2014 12:58

Georgy, the kind of pieces you referred to (flying through the air), are leapers. The Bishop is not a leaper, the main difference being that it observes the squares between its departure and arrival – while e.g. a (2,2)-leaper doesn’t observe square (1,1).

Georgy Evseev
Georgy Evseev
October 27, 2014 13:50
Reply to  dupont

My intention was slightly different. When, for example on an empty board Ubi-Ubi moves from one square to another, you cannot say which squares were visited and which were not. Still you know the position after the piece has moved. So, in fact, it does not matter if any squares were visited at all, only the result (position after the move) is significant. The same approach is in fact applicable to move of any pieces, including linear ones (and including long pawn move).

October 27, 2014 14:18

Indeed I have no example in mind of a fairy condition, where it is crucial to define e.g. a Bishop-move as a series of leaps between departure and arrival squares (and not by a single fly between those squares). But I will certainly not beg that it doesn’t exist!

October 28, 2014 15:35
Reply to  dupont

I think you were discussing about the possibility of introducing exactly such a fairy condition (or fairy piece). e.p. power !! Goodluck!

October 28, 2014 15:57

Good remark! Indeed Dominique and I are planning to construct a fairy condition where the difference between leapers and runners will be emphasized.

Georgy, this is true that in general we don’t know which squares are visited by ubi-ubi, it remains that some of them are really visited between departure and arrival!

You can see that by asking to Popeye which are the legal moves from a white ubi-ubi on a1, where b3 and c2 are occupied.

Stip ~1
Pieces white UUa1 black Pb3c2

1.UUa1*c2 !
1.UUa1*b3 !

In particular UUa1-d4 is illegal, which clearly shows that ubi-ubi doesn’t fly!

Dominique Forlot
Dominique Forlot
October 28, 2014 17:11

IAs the bishop and the rook slide on the chessboard and the slightest obstacle limits their moves, the Nightrider jumps up, with knight’s leap in knight’s leap, to his compartment destination unless it block on a occupied square before the goal; I think it’s the same way for all Leapers

That it is a question of sliding or of leaping the path must be exempt from trap.

That the considered piece is X or Y the graph of the movement has to be “in one piece”!

Nikola Predrag
Nikola Predrag
October 28, 2014 17:23

I wonder how seriously should I consider this discussion!
I’ve read the whole page again and I couldn’t find a single hint, let alone any reasoning, for explaining the crucial point.
And I don’t intend to count how many times I’ve asked the same question about it!
I have to assume that I’m too “blind” to perceive some self-evident definition in the standard chess rules, so that Georgy, Kostas and all others apparently can’t even guess what I’m asking for.

If I’m indeed so terribly “blind”, to miss something that is so obvious and self-evident to anyone else but me, I sincerely apologize for wasting everyone’s time and certainly my own, because I’ve learned absolutely nothing, despite all my efforts!

I still don’t understand what fantastic interpretation of the standard rules would allow anyone to ever think, even for a moment, that bPg4 might attack and capture wPh4.
I am sorry.

Georgy Evseev
Georgy Evseev
October 28, 2014 22:33
Reply to  Nikola Predrag

“A pawn attacking a square crossed by an opponent’s pawn which has advanced two squares in one move from its original square may capture this opponent’s pawn as
though the latter had been moved only one square. ”

Literally, this statement says that a black pawn, attacking square g3, may capture the white pawn, standing on g4, by making the move to g3.

Yes, pawn h4 captures pawn g4, while, to everyone’s shock., not attacking it))).

Nikola Predrag
Nikola Predrag
October 29, 2014 00:48
Reply to  Georgy Evseev

Which definition of a capture says that “when a piece moves to an empty square it can capture an opponent’s piece occupying the other square?
Standard rules clearly say that a piece moving to a square occupied by an opponent’s piece, captures the latter.

And perfectly logically, the ep. capture CAN HAPPEN ONLY ON 3rd RANK, because bP standing on 4th rank may execute a capture only on 3rd rank.
If bP captures wP on 3rd rank then wP is captured by bP on 3rd rank.
Where is the logic mystery or a hole in the rules?
Any other interpretation would require quite a convincing explanation to be worth even mentioning before it would be eventually rejected. However, I would enjoy in any interesting speculation.

“…Literally, this statement says that a black pawn, attacking square g3, may capture the white pawn, standing on g4, by making the move to g3….”
That’s an absolutely wrong conclusion.

– “this opponent’s pawn” –
determines WHICH UNIT may be captured and
– “as though the latter had been moved only one square” –
determines WHERE it may be captured.

A formulation of the rule combines the REAL and the HYPOTHETICAL (expressed by “as though”) elements.
– “may capture” means a REAL capture
– “as though the latter had been moved only one square” means a HYPOTHETICAL move

While we may speculate what could mean the word hypothetical or “as though”, the consequence of the rule is UNAMBIGUOUS and that makes the rule perfectly applicable and the logic consistent.
bP REALLY captures wP as though this wP is on g3 so, wP is REALLY captured by bP as though this wP was on g3.

All consequences of the REAL CAPTURE are REAL, and as long as they can be exactly and uniquely determined, the questions and answers about the hypothetical elements are irrelevant, since they are not needed in the reality.

The rule DOES NOT say “bP may move to empty g3 as though it captures on g3”
THAT would mean that “moving to an EMPTY square” is REAL (exclusively allowed, despite the general definition of pawns’ movement) and that the CAPTURE itself is HYPOTHETICAL.

Fairy chess might require the answers about those hypothetical elements but that is not a trouble of the standard chess.
Whatever some fairy condition changes in the standard rules, it all must be clearly defined.

Dominique Forlot
Dominique Forlot
October 28, 2014 18:21

I think that the pep is, in essence, “a fairy notion” in the standard chess rules!!

if, after 1.g2-g4, going back in time of a half-move is allowed for making a pep [1…h4xg3 ep.] I think we can accept, in case of rebirth of the g pawn, it occurs on g3 or g4 ( As we go back in time of half a move or more ) with the same fairy notion.

Even if the pawn h decides to capture it g3, the pawn g had decided to go upon g4 before the capture.

The final choice falls to the creator of the condition.

the catle is
in this position castle is legal.
1. 0-0!
where do the king crossed the rook?
They crossed on a square at some point… it is another “fairy notion” in the standard chess rules.

I am sorry if this explanation lands flat, It’s just my point of view.


Kjell Widlert
Kjell Widlert
October 28, 2014 21:48

I hesitate to make this enormous thread even longer, but let me add my two cents worth anyway.

Few things are completely logical in this world, not even in our little 64-square world…

Does a line move consist of several small steps, or is it one indivisible whole?
Consider the position wKa1 Rf6 – bKe8 Rh8 in orthodox chess. Castling is illegal, which implies that Ke8-g8 is considered to happen in two small steps Ke8-f8-g8, where the first small step is an illegal self-check (from Rf6).
Now interchange Ke8 and Rf6. The analogous move Re8-g8 is now legal, which implies (or at least might imply) that the rook is not travelling in small steps: then there would be a check after the first small step Re8-f8 and it would presumably be Black’s turn to move.

It seems to me most natural to say that a line-move is an indivisible whole, and that Re8-g8 never stops on f8, and that the Castling rule is one of those non-logical exceptions that make the game of chess more colourful or more fair.
This interpretation fits with the way most people (everyone?) interprets Madrasi: if we move bKe8>a8 in the first position above, Rh8-b8 is considered a legal move although it would be paralysed on f8 by Rf6 if it travelled in small steps (and would not be able to travel any further).

In ortodox chess, I think the only cases where it makes a difference whether a line-move is indivisible or a series of small steps, are the Castling and the check situations above, and the en-passant situation.

I was taught the en-passant rule the way that Kostas sees it: the P moves two steps, it really moves two steps, but the enemy pawn may capture it AS IF it had moved only one step. “As if” implies that it did not actually move one step, we just in some respects act AS IF it had moved one step. With that interpretation, it seems logical to me too, that a wPg4 captured en-passant is reborn on b5 (not b6) in Symmetry Circe.

Madrasi is a good example of a fairy condition where it matters how line-moves are actually executed. With “stepwise rooks”, bRh8-b8 with a wRf6 (and f7 empty) is definitely Madrasi-illegal. With “stepwise pawns”, Pg2-g4 with a bPh4 is Madrasi-illegal for the same reason (so we don’t have to worry about en-passant captures at all!).

October 28, 2014 23:00

Kjell, by asserting that “a line-move is an indivisible whole”, you just claim (or please tell me if I missed something) that runners don’t exist! Please have a look to the following link:


It doesn’t mean that Ba1-c3 is a couple of moves (Ba1-b2, Bb2-c3) where, say, the Bishop is able to check or to become paralyzed while standing on b2. It just means that Ba1 reached c3 after 2 leaps, and not after a unique fly…

Kjell Widlert
Kjell Widlert
October 28, 2014 23:42
Reply to  dupont

I didn’t mean to imply that riders don’t exist (of course they do!).

To use the illuminating railway metaphor from Fabel+Kemp that the text at the link refers to: a rook moving h8-b8 has to PASS the stations g8, f8, e8, d8, c8, but it doesn’t STOP at those stations, not even for a short while. For if it did stop at the stations along the route, we would have the effects of checking the other king, being Madrasi paralysed etc. So Rh8-b8 is not the move Rh8-g8 followed by the move Rg8-f8, etc, but an “indivisible” move Rh8-b8.

I admit that “indivisible” may not be the best term here: in order to state that Rh8-b8 is a legal move, we have to consider all the stations along the route and verify that it is possible to reach the destination by sequentially passing from one station to the next. The link ilustrates this clearly by considering more generalized routes than straight lines.

Nikola Predrag
Nikola Predrag
October 29, 2014 04:35

Chess rules had been so well defined that the game has survived a long history and that should be highly respected. The consequences of the moves are unambiguous despite our possible speculations about what exactly the movement of a piece means.

Any silly-looking interpretations may be considered as potentially valid if they come out of, and do not violate, these perfectly efficient rules.
The origins of chess suggest that it is an abstract imitation of a battle between two armies.
Quite logically, the rules do not suggest that the moves are “instant” and I don’t see anything that would clearly deny that the pieces move in time as the real units move on a battlefield.

The fact that, in principle, only one unit of one side moves while all the others appear as “frozen in time” during that move, indeed might look as “surrealistic”. But chess is a game and its specific concept of the time actually makes the chess so incredibly real-looking. Incredibly, considering the so simple rules.

In reality, when we notice some movement, our reaction can’t be instantaneous due to the inertia. And which rule says that it’s not the same in chess. One piece moves and the quickest reaction follows only when that move is finished. During the move, the moving piece does not interact with the “frozen” pieces.

However, there are the special moves, which apparently “violate” the general definitions of pieces’ movements. But they only treat the chess-time in a specific way, and since some default way is not exactly defined, there’s actually no violation of the rules.

In these specific moves, the pieces are exceptionally allowed to make more than one “movement” during one “move”. As a compensation for the opposing side, the opponent’s pieces are not completely “frozen” during such special moves, also exceptionally.

Ordinary “single” movement is quick just enough to be completed before the opponent’s reaction.
The execution of the “specifically combined” movements pretty logically takes more time than some “single” movement.
If a King is attacked on his initial square or while he passes across the attacked square and while the respective Rook also makes his movement, the opponent’s pieces might react and hypothetically capture the King. Of course, all these possibilities of both sides are precisely restricted.

What we might speculate about the specific rule of castling?
According to the general rules and detectable principles, a King may legally move away from an attacked square. So, he should be able to move from the attacked e1 to the attacked f1 and then to g1. If g1 is not attacked, when the castling is completed, the King would be not attacked.

Well, if that would be legal, someone could indeed speculate that the moves are instantaneous.
But it is not legal and the logic of principles could hardly justify the concept of “instant moves”.
The essential general principle is that a side in check can legally move own King to a square which is not attacked.

How a Rook can jump over a King during the castling, is a matter for speculations. It looks as a hole in the standard rules, at least as they are interpreted by FIDE for practical reasons.
Such movement of a Rook would violate the general rule about Rooks’ movement.
And in this case, there’s no mention of any “hypothetical elements” (like “as though”) which might virtually occur and then disapear as irrelevant when all real consequences of the move are determined.

However, some speculation based on the origins of chess, could be definitely rejected only if it somehow violates the standard rules, no matter how silly it might look.

Rook is a chariot, a King hypothetically could climb on it and step down on the other side. This can’t be done during an ordinary “single movement” while all pieces but one are “frozen” and can’t interact. During the castling, there’s time for the complex movements of partially “defrosted” pieces.

Silly? It certainly looks so and a contradiction with the rules should be easily spotted.
Until then, good night 🙂

Georgy Evseev
Georgy Evseev
October 29, 2014 09:15


The rules of chess are in fact specifically saying that en passant capture and castling are special cases and both have a special paragraph in rules describing exactly this case.

It means that a general approach is not available in case of these moves. The text of the rules concerning en passant is written so, that it is easier to understand what is going on, especially for beginners. Still, the effect is as I explained in previous message, starting with “literally”.

There is a lot of freedom of interpretation – especially taking into account a fairy region – but there is no implicit advantage in any specific treatment.

Moreover, chess composition is very much precedent-based. There is at least one rule which I consider wrong and there are several cases where I would have made another decision, if somebody would have been interested in my opinion. Still I follow suit in such cases.

(The example of me obeying “the wrong rule” may be found here: http://www.jurajlorinc.com/chess/ccm7tt_h.htm#uloha6)

Nikola Predrag
Nikola Predrag
October 29, 2014 18:50

that’s exactly the point, it’s ME who insists that the rules must be literally applied. So how the literal application of the same rule can be so different for you and for me?

“…The rules of chess are in fact specifically saying that en passant capture and castling are special cases and both have a special paragraph in rules describing exactly this case….”

I accept that, although I can’t actually find it in FIDE rules.

And what does that mean?
Special cases may be defined differently than the general rules as:
1. following some detectable general principle
2. contradicting the general principle

– Case 1. is pretty desirable since it confirms the logical wholeness of a system and makes a convenient basis for a possible development
– Case 2. is less desirable since the contradictions destabilize the logical wholeness and might heavily restrict a possible development

BUT ANYWAY, whether we like it or not, the rules have a purpose if they are applied.
Special cases explicitly define the special rule but ANYTHING that is NOT EXPLICITLY defined as a SPECIAL RULE, must OBEY the GENERAL RULES by default.
A special rule DOES NOT change all general rules unless it is explicitly defined so.

There is not even a hint that Special rule defines ep. capture on 4th/5th rank or that a double-stepping white Pawn could be captured on 4th rank by a black Pawn standing on 4th rank.
Or, that a Pawn “may move to a square NOT occupied by an opponent’s piece, which is diagonally in front of it on an adjacent file…”. It’s not stated “may move to an empty square as though it’s occupied by a Pawn which occupies another square” or any formulation that might justify the similar speculations.
Where is this ever mentioned in any paragraph of the rules? What might be called as “literally interpreted”?
“Literally” – for something not written or spoken or expressed anyhow??!!

I’m interested in various interpretations and possible logically consistent alterations of the standard rules.
Why is it so hard to agree about a literal meaning of a few sentences and particularly about what is CLEARLY NOT a literal meaning?

It seems that the practice is considered as a literal expression of the rules. If in practice we first move bP to g3 and then remove wP from g4, this “should” be literally understood as though the ep. capture was made on g4.
That would be an completely inverted reasoning.

The rule says that bP may move to g3 if it can capture on g3.
The capture literally HAPPENS “as though” on 3rd rank, and it is stated which pieces are affected.
And this special rule retains the general principles of standard rules, so no wonder that it was formulated and written as it is, not only for the beginners.

If A affects B through some procedure X, then B is affected by A through the procedure X.
That’s literal and if any other interpretation could be possible, it would require a convincing explanation.
So, wP is literally captured on g3.

I certainly don’t insist on the rule as it is written in FIDE-rules, but I have never seen anywhere the en-passant rule formulated as bP captures wP on g4 and ends the move “as though” on g3.

Georgy Evseev
Georgy Evseev
October 29, 2014 19:56
Reply to  Nikola Predrag


The fact is that capturing piece arrives to g3, while captured is removed from g4. These two squares are _different_ and this is what defines en passant as a special case. I would dare to say that in this case “square of capture” is not defined at all.

So, as with my Locust example which you did not comment on, the rebirth square is determined by _square of removal_, and not by _square of arrival_.

Nikola Predrag
Nikola Predrag
October 29, 2014 23:07
Reply to  Georgy Evseev

What “literally” means to you?
The square of capture is not explicitly and literally noted.

It is explicitly stated that:
1. bPh4 may capture wP as though wPg2 has MOVED to g3.

Since nothing else is stated in the specific rule, we must literally apply the general rules to complete the move.

So, how else might be the general rules literally applied but “as though” bP has captured, and wP was captured, on g3.
What might suggest a literal meaning that wP was captured on g4?

What in the rules says that a piece is captured by a hand and not by an opponent’s piece?
And what explicitly says that during the capture wP occupies square g4?
There is NOTHING that could literally mean that wP is captured on g4 and the removal by a hand is only a manual adjustment of the “material” position to match the true chess-position after the capture on g3.
The true chess-move is defined “as though wP has MOVED only one square”.

The chess rules are defined in an “abstract system”, they do not depend on material features.
Note that wPg2 can be MOVED TO THE SQUARE g3 but nothing says that it can be MOVED from g2 TO THE SQUARE g4!


All other moves are defined as “a piece moves TO A SQUARE” which literally means that the mentioned piece occupies a particular square when the move is made.

Literally, it’s not defined which square a Pawn occupies after the “two-squares advance” since it was NOT literally MOVED TO ANY PARTICULAR SQUARE.
The generally defined “one-step move” on the first Pawn’s move (from the initial rank) has an alternative, to “advance further” but not literally ending a move onto some explicitly stated square.

I think that g2-g4 means that wP has moved towards g4 but can’t completely arrive to that square before bPh4 decides “to capture or not to capture” it.

I think that wP can’t be “literally” captured on, and thus removed from, square g4. Nothing explicitly defines that wP is captured on g4, therefore this can’t be a literal meaning.

And I think that literally means “literally” and not “something I believe, no matter what is explicitly stated”.

What about the Locust example?
Should I comment how the standard rules hypothetically explain it or the opposite?
I’ll try the first question.

When piece X moves to a square A occupied by some opponent’s piece Y, X captures Y.
So Y is captured by X on A.
If X is a Locust, X moves to A, captures Y and continues to move to a square B which is specified by the definition.
Again, Y is captured by X on A.
In case of AntiCirce, Locust X arrives to B and then is reborn according to square A.

The Cheylan type does not seem precisely defined. It could be: Capture is not allowed if the rebirth sqare is the characteristic arrival square of the capturing piece. Then the Locusts could capture on the rebirth square.

October 30, 2014 02:35

“I think that g2-g4 means that wP has moved towards g4 but can’t completely arrive to that square before bPh4 decides “to capture or not to capture” it.”

This is precisely my “convention B”, see the message october 24, at 19:57. And indeed I think this is the best way to handle the precificity of the ep move.

Georgy Evseev
Georgy Evseev
October 30, 2014 10:35

“What might suggest a literal meaning that wP was captured on g4?”

Easy, the captured piece is removed from g4. Look how the en passant capture is played in any real otb game, and it will be obvious for you.

As I said before there are two things really happening during en passant capture: 1) black pawn moves from h4 to g3; 2) white pawn disappears from g4. Everything else is interpretation.

My approach is direct: what _has to happen_ is exactly what happens.

Your approach is different: you add an additional step of returning of a pawn to g3 to make a “traditional” capture. This is an interpretation and in reality there are no logical arguments which can force to consider any interpretation right or wrong.

You are also probably right that there some “unnatural things” happen when my approach is used. But you do not see that other “unnatural things” happen when your approach is used. For example, I very much dislike the idea that same piece in same position may be captured in different squares; I also do not like the concept of moving enemy piece before capturing it (and you move it from g4 to g3; or to g2 and back to g3) or an idea of “unfinished moves”.

Dominique Forlot
Dominique Forlot
October 30, 2014 13:25

“Where goes the ghost?” : this is the only question we have to answer!

Perhap , must we define what “type of pep” existe on each “Fairy Condition”

“The Lilliputian religion says an egg should be broken on the convenient end, which is now interpreted by the Lilliputians as the smaller end. The Big-Endians gained favour in Blefuscu.” Jonathan Swift

“pep type I” : “Small pep” = Lilliputian’s Type( breaking the egg on the Smaller end)

“pep type II”: “Large pep” = Blefuscu’s Type( breaking the egg on the Larger end)

we have now a “Small pep”, where the ghost piece goes in underworld on the square where the capture arose and an “Large pep” where the ghost piece goes in underworld on the square reached before the capture.

here is only what we need ( after love, indeed!)

October 30, 2014 16:43

In e.g. math, when we have to face a difficult question, the reflex is to imbed it into a general theory, enough strong to be able to provide the answer. As already mentioned, Dominique and I are planning to define what we are calling “ep-power” with the scope, among others, to be able to analyze the orthodox ep-capture as a particular case.

The central idea of what should be a piece equipped with ep-power (ep-piece for short) is clear: each piece crossing a square observed by an ep-piece from the opponent side may be captured on this square by the ep-piece.

As an example, consider the following position:

W: Kg8 Ba1,
B: Kg6 Qg7 Pa3, this latter equipped with ep-power.

The only possible white move is Bxg7 and it would be quite absurd to declare this move illegal, i.e. that white is checkmated in this position. But now any reasonable definition of ep-power should allow Pa3xb2-ep, with annihilation of Bg7.

The conclusion of this speculation is that Ba1 really reached square g7 during the ep-process, as something strong happened on this square – the capture of the black Queen. Now if we apply this enlarged ep-theory to the orthodox case, I have to change my mind and recognize that g2 really moved to g4 during the sequence g2-g4 hxg3-ep.

If we accept this viewpoint, the fairy element in the ep-process is that a capture occurs on an empty square – the captured unit standing elsewhere. This is a rather high prize to pay but the alternative – white is checkmated in the above position, is even more difficult to accept.

Nikola Predrag
Nikola Predrag
October 30, 2014 22:51

Well Georgy, I can’t follow your understanding of literal application of the rules.
You base your “easy understanding” on:
“…Easy, the captured piece is removed from g4. Look how the en passant capture is played in any real otb game, and it will be obvious for you….”

My own “obvious interpretation” is that OTB players play according to the previously defined chess rules and not that the players define the chess rules.

So, I hope that we are discussing about the rules by which the players play.
And none of the standard chess rules says that wP should be removed from g4. If the players obey the rules, they literally remove that wP FROM THE BOARD, but NOT FROM g4! The removal from the board is allowed by the rules, and the removal from g4 is NOT allowed!

wP can be removed from the board only HYPOTHETICALLY “as though” it was occupying g3 during the ep.-move.

If you can’t accept that the reality might be a consequence of a hypothetical features, OK, then the ep. capture is not real but only hypothetical, and wP will remain on g4 in reality. Nothing would cause its removal. And bP would not be able to move to g3 in reality.
And since ep. capture would not happen in reality, the move by Black would NOT be really played and Black would have the move since nothing has really happend.

But you may try to explain what in the rules not only allows, but even requires the removal of wP from THE PARTICULAR SQUARE g4.
I am able only to understand that the rules allow to remove wP from THE BOARD “as though it was occupying g3”.
And accept it as reality, despite the “obviously” hypothetical features.

It is perfectly obsious that the Earth stands still and that the Sun moves around it. Don’t listen to those who say “eppur si muove”.
Listen to the otb players who think they remove wP from exactly g4, not just from the board.

Georgy Evseev
Georgy Evseev
October 30, 2014 23:13
Reply to  Nikola Predrag

“The proof of the pudding is in the eating”. I refuse to further build Babylonian towers where everything can be explained much easier. Consider yourself winning this in this feast of silliness.

Nikola Predrag
Nikola Predrag
October 31, 2014 03:37

Yes, I’m silly enough to see a difference between “apparently” and “literally”.
For the “literally case”, it should be found in the rules where the removal is mentioned:

“3.1 It is not permitted to move a piece to a square occupied by a piece of the same colour. If a piece moves to a square occupied by an opponent’s piece the latter is captured and removed from the chessboard as part of the same move. A piece is said to attack an opponent’s piece if the piece could make a capture on that square according to Articles 3.2 to 3.8.”

The relevant extract:
“If a piece moves to a SQUARE OCCUPIED by an opponent’s piece the latter is CAPTURED and REMOVED from the chessboard as part of the same move.”

It’s very, very, very easy to understand WHEN a piece can be removed from a chessboard.
What the otb players think when removing a piece, is perhaps interesting to a psychologist but not so much to me.

Perhaps they build the Babylonian towers to explain and justify the appearance.

I don’t see that a “much easier explanation” would be:
“a piece captures by moving to an empty square and the captured piece occupying another square will be captured and removed exactly from that another square”

Quite naturally, I can’t understand that easiness, due to my silliness.
Maybe some day I’ll understand the “true” reasoning.

“The proof of the Sun moving across the sky is in the observing (certainly not in the laws of physics)”

And when I eat the pudding, I’ll might see that
“observing the otb players proves that wP is not only removed from g4 (which is so clearly obvious) but that wP is also captured on g4.”
And the square of capture determines the rebirth square, as the rule above the problem No.627 so nicely defines.

October 31, 2014 18:10

“a piece captures by moving to an empty square and the captured piece occupying another square will be captured and removed exactly from that another square”.

Indeed this is close to the definition of the ep-power (with the addition that the empty square has been crossed by the captured piece) – as Dominique and I are working on!

EP-pieces will therefore constitute a new family of fairy pieces, with capturing rules not stanger than e.g. for Locusts.

Nikola Predrag
Nikola Predrag
October 31, 2014 19:19
Reply to  dupont

To make a (standard)-capture, a piece must come to a square occupied by an opponent’s piece, then it can proceed to some other square (eg. if that capturing piece is Locust.

Perhaps that could be the essence of standard-captures: an interaction of opposing pieces which hypothetically both occupy the same square in some moment.

Distant interactions, (as in Madrasi but also as the standard meaning of attack and check), are not captures, of course.
But if some distant interaction may be treated as a capture, without a “crash” of both pieces on one square, that might be a type of “non-standard-capture”.

“X captures Y on square A, so Y is captured by X on square B” – this might be a logic-trap, even if could work unambiguously in practice.
It would be nice to define the nature of such distant interaction, if possible.

The ep.-pieces could have the power to pull back the “crossing piece” after it has finished a “characteristic part of its move”, as in your example:
1.Ba1xg7-b2 axb2ep.
That would be a “standard-capture” on b2.
Of course, all depends on your intentions.

October 31, 2014 20:31

For sure, something has to be non-orthodox in the definition of ep-power, or ep-piece – as it is already the case in the standard ep-capture.

The way you are thinking, if I understand it correctly, is that this necessary non-orthodoxy should nevertheless follow some classical rules – which might be called “universal laws”.

I fully agree on that point, fairy chess should not be the land where everything becomes possible – some basic structural elements have to be keep.

Locust has the particularity to move after a capture (like any piece in T&M) – this is very fairy. And you’re right (excellent remark), one can view ep-pieces as able to pull back crossing pieces – very fairy too, but not more to my mind.

The implication of this nice viewpoint on the standard ep-capture is exiting and maybe new: wPg2 really moves to g4 (nothing fairy at this step) but bPh4 – which is an ep-piece regarding Pawns, has the power to retract wPg4 to g3, and to capture it on this square!

“Hability to pull-back” would then be the only fairy element in the whole ep-procedure, a prize which seems reasonable to pay regarding other possible interpretations.

Dominique Forlot
Dominique Forlot
October 31, 2014 21:10

We can see also the ep- piece as an archer positioning on the square of crossing and aiming at the opponent at distance.
Here, we can imagine the bishop staying on b2, and his arrows pierce throw g7: no need to drag the queen’s body on b2.

There is no denaturation of the game of chess which this kind of fairies. Logic and rigour can meet imagination and creativity.

Would love your thoughts, please comment.x