Fairy chess composition

# No.1739 (MR)

Original Fairy problems
JF-2023:
01.01.2023 - 31.12.2023

☝ Definitions

No. 1739 Manfred Rittirsch
Germany
original - 31.03.2023

white Re1 Pf2c2h3g3e3f7e7 Kh4 Bc6 black Sd1 Kd2 Pa2c3h5 Ba8
hs#2            b) Ph5->g5            10+6
Isardam + Mars Circe

Solution: (click to show/hide)

Subscribe
Notify of

Inline Feedbacks
seetharaman kalyan
March 31, 2023 23:41

Unusual combination of Mars with Isardam. I agree that observation should be linked to capturing possibility. Quite complex effects in two moves.

Joost de Heer
April 4, 2023 17:33

Popeye uses the orthodox observation lines for Isardam I think, not the Mars-Circe lines.
Compare with a composition I published in feenschach almost 20 years ago.
W: Ke8 Pb6
B: Pe2
h=2**
MarsCirce + Madrasi + No capture
1… b7 2. e1=Q b8=Q=
1… b7 2. e1=R b8=R=
1.e1=B b7 2. Bd2 b8=B=
1.e1=S b7 2. Sf3 b8=S=
(no capture to avoid 1.e1=B b7 2. Bc3 bc3=)

Last edited 1 year ago by Joost de Heer
Joost de Heer
April 7, 2023 14:26

Oldest composition I can find with Madrasi/Isardam + MarsCirce is a h==2 by Roméo Bédoni from 2001.

Manfred Rittirsch
April 11, 2023 10:32

Thanks for the input, Jost! The minimanner problems by Bédoni and yourself are nice pioneer problems for the use of observation lines connected to home squares. Nevertheless the term „Madrasi/Isardam” is misleading, because these conditions are substantially different and should NOT be treated as a couple. In fact there is nothing more deviant from Madrasi than Isardam.

Walter Lindenthal
April 5, 2023 19:37

I think it’s a matter of timing…

Sorry my BIG mistake, I removed my original comment! Either white (defensive) move after a) 2…a1=Q or b) 2…a1=R would actually create an Isardam situation (established by the Marscirce mechanism)! Especially in a) the wrongly suggested defensive move 3.wQf8-f6 would just change from Qa1xQf8 (via d8, in case wSf8 moves) to Qa1xQf6 (via d8). Unfortunately I overlooked that Qf6 would equally be observed by Qa1 under Marscirce, I only concentrated on defending the wKh4! I now fully agree, the solutions are correct and really great! My apologies for this confusion!

Last edited 1 year ago by Walter Lindenthal
Walter Lindenthal
April 7, 2023 18:09

A really fascinating problem that gave me lots of thinking. Even the final “forced” black move under the stipulation hs#2 (incl. the kind of different promotion in a and b) is defined by the effects of these two conditions! As far as the difference between Popeye and Winchloe is concerned, this might be based on a different “priority”-interpretation (see feenschach TT-80 “Priority”)! If Isardam has priority over Marscirce, then an observation from the (interim) rebirth square of Marscirce (bQd8, respectively bRh8) will immediately fall under the Isardam condition – this seems to be the given interpretation in the solutions here. I fully agree with that!

Theoretically the alternative would be that Marscirce has priority over Isardam (i.e. the observation issue should only be considered once the Marscirce capture move has fully been executed). Here this should result in a very different outcome, that is to say in a) 2.e8=S# or in b) 2.f8=B# which means checkmate of the bKd2, because Marscirce (and the checkmate by Marscirce) has priority over Isardam. But actually this might hardly fit with the definition of Isardam as “…ANY move, including capture of the King, is Isardam illegal if a Madrasi-type paralysis would RESULT from it…”. Hard to say, anyhow the situation is completely different with other combinations like Isardam+Masand, etc… where two different priority-interpretations are clearly feasible…

***Edited*** I found a good example for priority of Anticirce over Isardam from Eric Huber at yacpdb 362029…

Last edited 1 year ago by Walter Lindenthal
Manfred Rittirsch
April 11, 2023 12:42

Please allow me to share my opinion on two subjects.
The first is the definition of my favorite fairy condition Isardam. I understand that the inventor used the Madrasi condition to define it, which initially made sense, because it shed light on the name’s origin, but that Isardam definition does not stand the stringency test because it depends on another one. That’s why I prefer to slim it down to make it refer to the aspect of observation only (which is “by chance” the basis of paralysis in Madrasi). This simplification was obviously taken into account in the Fairy Chess Classification Project and WinChloe (among other places), too.
(Btw. I just noticed that the definition of Isardam given with the Fairy Chess Classification Project is wrong in a different respect:
“A move or capture may not be made if it leaves two opposite-colour units observing one another. This applies to checks also, so that a check is not effective if the threatened king capture would leave two opposite-colour units observing one another.”
It is NOT required that two opposite-colour units observe each other. Note that there are many cases where the observation is not reciprocal, e.g. when using Grashoppers: with wGa1, bGa4, Pa3, Ga1 observes Ga4, but not the other way round. This asymmetry was already examined during the Andernach meeting 1986, see e.g. YACPDB ID 304493. This one-sided type of observation is also forbidden in Isardam.)
When it comes to the combination of fairy conditions, I see the general importance of prioritization, and I guess we all agree that this makes sense only in those cases where it makes a difference. But there is also another aspect on which the programmers of WinChloe and Popeye (among others) seem to disagree. This aspect shows up when you try to consider all applied conditions “at the same time”, which in my opinion should be the default approach whenever it is possible to assume a natural prioritization. In my opinion this is the case with Anticirce + Isardam as well as with Mars Circe + Isardam. My argument is that rules affecting the capture naturally precede rules affecting observation, as observation is based on capture. On the other hand, in specific situations the possibility to capture may depend on observation according to Madrasi or Isardam, but NOT the rules themselves (requesting a rebirth according to Anticirce or a modified starting point for captures in Mars Circe). The related considerations are made on different levels of analysis. That’s why I think that for these combinations there is no need to add an explicit prioritization rule unless you explicitly want it the “unnatural” way. I am already planning an article on that discussion.

shankar ram
April 12, 2023 07:58

Thank you for pointing that out, Manfred! We will replace “two opposite-colour units observing one another” with “a unit observes/is observed an/by an opposite coloured unit of the same kind” in the definition.

Or maybe the definition in Julia’s original fairy terms database (actually used here for this problem!) is even better: “Any move, including capture of the King, is Isardam illegal if a Madrasi-type paralysis would result from it”.

The related condition “Leffie” does not have this issue! 😉

Last edited 1 year ago by shankar ram
Manfred Rittirsch
April 12, 2023 18:21

Please read again the beginning of my comment above to understand why I definitely prefer your first proposal!

Joost de Heer
April 12, 2023 19:27

Opposite coloured seems to be inexact as well, since it disregards neutral pieces

Walter Lindenthal
April 12, 2023 12:58

I am glad the opportunity allows me to debate an issue that has been burdening my thoughts for some while… Yes, I do agree, the definition of “observation” is the central factor here, and it is based on the idea of capture…, but what kind of capture? “Natural” capture as unimpeded by any other condition (like Marscirce), or capture in the specific way as defined by Marscirce (rebirth “before”) or by Anticirce (rebirth “after”)…? The definition of observation (“Beobachtung”) in Maerchenschachlexikon of Schwalbe seems to basically EXCLUDE “restrictive fairy conditions” when it comes to the identification of observation (unless Marscirce would not be understood as a “restrictive fairy condition”). But on the other hand, if you DO exclude them, how can the interim rebirth square (of Marscirce), or an after-capture rebirth square (of Anticirce) be of any influence when it comes to the detection of observation (for Isardam)? As far as I know there is even a rule in Anticirce referring to a capturing King: a check of the King never can occur on the capture square, it can only occur finally on the rebirth square of the capturing King…!? So, does the combination of Marscirce+Isardam create a “double standard” for observation? As follows, a simple example: wQa3, bQe3. Do these positions establish observation under Marscirce+Isardam???

a) YES(?), you must apply a kind of “natural” observation (by the inherent move capabilities of a stone) WITHOUT taking into consideration that moves of capture (as a prerequisite(?) for observation, and only as such in turn creates the Isardam restriction) would require the application of the Marscirce rebirth mechanism here. The positions would then be illegal because of “natural” Isardam…

b) NO(?), but then you have to ignore the definition in Maerchenschachlexikon (which seems to exclude Marscirce as a “restrictive fairy condition” when it comes to the identification of observation). This should mean for the given positions in this simple example, under the Marscirce condition there is no capture possible (but which would establish the necessary precondition for observation in Isardam). Thus, the positions are legal, and consequently 1.wQa3-b3 should equally be a legal move. But what about the continuation 1…bQe3-e1? If the “natural prioritization” of Isardam applies, the move should be illegal, because it creates (one-sided) observation through the Marscirce rebirth mechanism (via d1)! But in case an “artificial prioritization” of Marscirce over Isardam would apply, this should result in a logic dilemma (virtually!): If you prioritize Marscirce in the evaluation, then once an immediately following capture move under Marscirce 2.wQb3(via d1)xbQe1 will be executed, there is no black piece left which could establish an Isardam situation (in a secondary evaluation)! But would instead an alternative move 2.wQb3-c2 equally be legal (a one-sided observation had already been established before!), this should actually create an illegal Isardam situation..? And the more an alternative move 2.wQb3-d1, because now you’ve even got a two-sided observation…

So, which rules have to be followed, when the possibility to capture under Marscirce depends on (non)observation? It seems to be some kind of circular reasoning… I am really confused… THANKS!

***Edited*** I added the word “(virtually)” for clarification…

Last edited 1 year ago by Walter Lindenthal
Manfred Rittirsch
April 14, 2023 14:39

Yes, you are trapped in a vicious circle of reasoning! To get out of it, just imagine a world where the laws of gravitation (~ Mars Circe) and the laws of (anti)magnetism (~ Isardam) apply at the same time, and then replace one of the Queens in your example by a Rook. I hope you will get the (or maybe my?) unambiguous general concept of capture & observation then. Here the possibility to capture does NOT depend on observation, but a) on a specific spatial relationship between the involved pieces which is NOT an observation (yet): the location of the piece to be captured must be on a square that can be accessed by the (still virtually) capturing piece – and b) the feasibility of that capturing move including all the required consequences (like mandatory rebirth). Only when both prerequisites are meet, I call this an observation.
I guess the definition of observation (“Beobachtung”) in the “Märchenschachlexikon” refers to (usually temporary) restrictions of the capabilities of certain pieces in specific situations (positions), like the inherent priority rules of e.g. Maximummer (“Längstzüger”) or Single combat (“Duellantenschach”), but NOT to general restrictions/modifications imposed by a condition (otherwise I would strongly disagree).

Walter Lindenthal
April 14, 2023 18:11

Thanks! I would agree with your interpretation of the definition of observation in Maerchenschachlexikon. May I try to redefine your “spatial relationship which is not an observation (yet)” with the idea of a “directed kinked ray” from the position of the one piece through its rebirth square (Marscirce), and on from there to the position of the other piece strictly following its inherent move capability (to be established for both pieces counter-directionally). If one or both ray(s) hit their target then it is observation, and the move would be illegal under Isardam if an opponent piece of the same kind is hit (else the move would be legal). This ray concept could abstain from the direct notion of “capture” and as such sidelining the idea whether capture (of Marscirce) or observation (of Isardam, based in circular logic on capture!) has priority…

For my simple example under Marscirce+Isardam with wQa3 and bQe3 (if I replace just one of them by a Rook I woun’t have an Isardam situation, so I have to replace both of them…) this would show: Neither the starting position wQ/wRa3, nor the move 1.wQ/wRa3-b3 would establish a ray that hits bQ/bRe3, because under Marscirce it would have to go through d1 (for wQ), or h1 (for wR), or for the other side through d8 (for bQ), or h8 (for bR). So there is no observation yet. But the move 1…bQ/bRe3-e1 should be illegal under Isardam, because now the ray hits (wQ via d1 / wR via h1)! Is this equivalent to what you meant?

The idea with the ray should also be applicable with (e.g.) a capture under Anticirce+Isardam. The ray goes strictly following its inherent move capability from the capturing stone to the captured stone and from there to the rebirth square of the capturer. If the priority is with Anticirce then the (possible) observation is only realized on the rebirth square, if the priority is with Isardam then the (possible) observation is already realized on the capture square… Both of these squares are defined only AFTER the capture, so capture / rebirth / observation can clearly be disconnected logically from each other (which is a fundamental difference to an ADVANCE-rebirth of Marscirce+Isardam)…

Manfred Rittirsch
April 18, 2023 23:13

I explicitly wanted you to replace only one Q by a R to illuminate the different outcome of that “spatial relationship” when the pieces are different.
On a 1st level, there are “replacement” capabilities based on the specific scope of a figure (here orthogonal straight sequences of adjacent squares called “lines” for R and additional diagonal straight lines for a Q. For all orthodox pieces excepting pawns, in orthodox chess the same scope applies for capturing and non-capturing moves. There is only one additional requirement for a capturing move: a piece of (usually) different colour must be on the destination square (or another specified square, e.g. the nearest adjacent square for a Locust). In Mars Circe, however, the capturing and non-capturing scopes of Q or R differ with regard (only) to the starting square. In a constellation like wQa3 + bRe3 the pieces are blocking each other (with or without additional fairy condition Isardam) so that the wQ cannot reach e(!)-h3 and cannot capture at all, whereas with a bR e.g. on d4 the wQ can move everywhere on the a-file or 3rd rank, but capture only on d4. The Q IS already on d1 when “leaving” for a capture. There is NO movement to an interim square and no “kinked ray”, it’s all based on (here: separate) scopes.
Assuming that observation is based on the possibility to capture, the capturing scope must be recognized before any observation check, which consequently obtains the status of a 2nd level consideration. A piece observes another one from the starting point of its capturing scope, and you must not position any piece in the capturing scope of a similar piece in Isardam. This is valid for Isardam alone or in connection with Mars Circe or Anticirce (see e.g. P1246071, P1256271).
Compare to the 2nd level consideration of guarding (… an own piece in the capturing scope) in https://juliasfairies.com/problems/jf-2013-ii/no-324.

Walter Lindenthal
April 16, 2023 09:28

If I may add a last thought, by example: Marscirce+Isardam, wQa3, bQh4, bRh1; 1.wQa3(via d1)xRh1. Both Queens do NOT observe each other, neither on starting square a3 nor on capture square h1. Is this sufficient for the move to be legal under Isardam (if so, you could call it “priority=Marscirce”)? BUT the bQh4 (in Marscirce-style via her interim rebirth square d8) DOES observe the interim rebirth square d1 of this w move (you could call this in comparison with other types of combinations as “priority=Isardam”). Which one is the “regular” way to understand the Isardam situation, and are both of them valid interpretations (possibly resulting in different software implementations)?
The answer should be a necessary input for further understanding the Isardam problems that arise from Marscirce-variations (with advance-rebirths!) of this combinations with Isardam, like with MarsCuckooCirce, or instead with Circe Jovien, or Circe Jovien Cuckoo…, and especially when including fairy pieces (variable rebirth squares depending on position!) – THANKS!

Georgy Evseev
April 16, 2023 11:12

I’ll add some matter for thought.
What is the “move”? My interpretation is that this is an instant conversion of position from one state to another. So, imtermediate states “inside the move” are _not_ to be taken into account. This, of course, directly contradicts the idea of this problem.
A number of examples to support this notion.
W: Qd1 – B:Qb5, Isardam. 1.Qd8 is legal, there is no “crossing” of squares d3, d5, d7 controlled by BQ.
W: Se2 – B:Kb5 Sd4, Take&Make. 1.Sxd4-c6 is OK move, there is no “intermediate check” while WS is on d4.
W:Ka1 pa4 – B:Ga8 pb3, Mars Circe, Isardam. 1.a4-a2xb3 is perfectly legal, there is no “illegal self-check”. Why the situation is to change if we put Ga1 instead of king?

Walter Lindenthal
April 16, 2023 14:00

Good thinking and helpful contrast in order to further a clarification, indeed is “the move” defined anywhere at all!? I am not sure about the (formal) correctness of seeing it by all means as an indivisible unit – compare with “priority” ideas (at last feenschach TT-80)…
1st ex) agreed, but here is no “defined” interim square to be respected, like with rebirth rules, take&make, etc… only one “indivisible” move…
2nd ex) I am not sure if NOT checkmating the King here is allowed at all, and instead move to c6 …? Shouldn’t the starting position wSe2 already establish a checkmate against the bKb5 under take&make?
3rd ex) I assume the wPa4 is already in an illegal Isardam position with bPb3 under Marscirce? So, let’s exchange bPb3 with bBb3. In your example it’s the hurdle that moves, it’s not an “attacking move” of the G! At the end of the hurdle’s move there is no hurdle for the G any more to capture the K. And as far as I know there are rules regularly saying that checkmating the King only is in effect at the end of a “compound” move (with only a few defined exceptions)…
But I am not sure about the sub-example (exchange wKa1>wGa1) here! Maybe the same reasoning, it’s none of both Gs that move, and the hurdle must finish its move first before you can judge an Isardam situation between the Gs. That seems to be rather different from a situation where (theoretically) one of the Gs is moving and not the hurdle…?
But I do NOT “know” the answers, it’s just my feelings…

***CORRECTION***
2nd ex) ***sorry, my mistake*** – the make-partmove must not capture! Consequently there cannot be an “intermediate check” while wS is on d4, this is part of the t&m-rules.

Last edited 1 year ago by Walter Lindenthal
Manfred Rittirsch
April 18, 2023 23:18

wPa4 + bPb3 is already illegal in Isardam, because square b3 is within the capturing scope of Pa4, so there is an observation of Pb3 by Pa4 (not the other way round).

Joost de Heer
April 19, 2023 16:57

I think this is the mathematical representation of Madrasi and Isardam:
Let X be a position, and S be the side to move.
Let P(X,S) be the possible moves in position X with side S to move (i.e. the moves that a move generator would generate for a given position with a side to move).
Madrasi: If P(X,S) contain a move in which a piece(1) of type A captures a piece(2) of type A then piece(2) is Madrasi-paralysed.
Isardam: To determine whether a move p from P(X,S) is legal, calculate the possible moves P(Y,S) for both sides, where Y is the position after move p. If this collection of possible moves from both sides contains a move in which a piece of type A captures a piece of type A, then the move p is illegal.

Kevin Begley
April 19, 2023 21:01

I would go one step beyond what Joost says.
Let X be a position, let S be the side to move, let C be a list of fairy conditions (rules).
Let P(X, S, C) be the possible moves in position X with side S to move under a list of conditions (rules) C.
Some people seem to think Isardam/Madrasi can ignore the rules of the game (the conditions) when calculating whether a one piece (piece1) of type A can capture another piece (piece2) of type A. Neither Madrasi not Isardam are conditions which explicitly state that this calculus is based upon orthodox rules. Therefore, we should all reject that nonsensical interpretation. The capture potential of two like units must always be based upon the rules of capture (unless specifically stated).
Around 2014, I became aware that Popeye has a bug when Isardam is added to some conditions (I know some folks will not like the fact that Win Chloe gets this right, generally, and Popeye is wrong; while I can appreciate their instinct to trust the freeware interpretation, they should be aware that a member of the Popeye team admitted this bug to me back around 2014).
That year, I had composed a 4-man Babson, using Anticirce Equipollent + Isardam.
Somebody reported finding that my problem revealed a discrepancy between Popeye and Win Chloe (it is highly likely I found and reported the bug, but I can not recall).
As I recall, a member of the Popeye team contacted me, admitted this bug (he said that Popeye should not be trusted when Isardam is combined with any other fairy condition), and he very graciously offered to make immediate repairs (so that my problem might compete in a certain 4-man tournament).
I am confident that my memory is correct here, because I was not even aware, at the time this offer was extended to me, that this tournament (which I once esteemed highly, which I specifically had hoped to enter) would exclude my problem on the basis of a Popeye bug.
Not only did this event shatter my faith in problem chess, it also put me at the crossroads: On the one hand, I specifically wanted my problem to compete in this tournament, and on the other hand I could not in good conscience participate — as a beneficiary, no less! — in the culture of rampant (yet subtle) favoritism. As I recall, I went out of my way to downplay — if not outright discourage — the remedy offered to me. It seemed grossly unfair for me that I should explicitly encourage the urgency of that offer (which would confer some advantage to me).
I do regret that I did not follow up on this offer — this debate might have been averted — but in my defense, let me say this: I quit problem chess (for several years), not long after this experience, and this faith shattering event was one of many significant factors.
Manfred has it right. Win Chloe has it right. People inclined to favor whatever interpretation Popeye offers should be aware that a prominent member (at least one) of the Popeye team did admit (more than 8 year ago) the tool is not correct, and never trustworthy when Isardam is not the only fairy element involved.
I’m sorry I didn’t follow up on this offer to repair Popeye (after I was no longer a direct beneficiary of potential favoritism). This whole debate might have been avoided.

Walter Lindenthal
April 20, 2023 11:24

Ok, but P(X,S) / P(Y,S) do NOT specify the specific kind of “capture move”! Are there only two marker points connected between starting – capture squares, or three (or possibly even more) marker points connected between starting – (advance-rebirth/e.g. Marscirce) – capture – (after-rebirth/e.g. Anticirce) squares? As Kevin wrote (quote): “Neither Madrasi nor Isardam are conditions which explicitly state that this calculus is based upon orthodox rules… The capture potential of two like units must always be based upon the rules of capture (unless specifically stated).” So, in P(X, S, C) the C must – if applicable – encompass the specific capture rule for Marscirce (advance-rebirth) or Anticirce (after-rebirth), and IMO it’s important to clearly define its consequences for the capture-based Isardam interpretation (i.e. which “marker points” are to be respected)…

Manfred Rittirsch
April 20, 2023 17:51

Please see my reply from April 18 to your other post above to evaluate the concept of scopes!

Walter Lindenthal
April 21, 2023 14:13

THANKS for your efforts in trying to clarify this issue, I really do appreciate it. But unfortunately I cannot follow your explanations (although I feel quite comfortable in Circe-issues in general)! I apologize, but I do not understand your point, nor the rather ethereous expression “spatial relationship” and your repeated use of the term “scope“. They do not give me any hints for how to nail the issue down in specific details and whether pieces are observed on which “involved” squares and under which framework rules. Or may be there is a basic misunderstanding? [Remark: this is my reply from yesterday which got censored somehow…!]

In the combination Marscirce+Isardam there is a double use of the idea of “capture”. Marscirce defines the capture move in a specific way (“via” rebirth square of the capturer, or you could equally say “starting” from the rebirth square). And Isardam is based on observation which in turn is based on the idea of capture, too. For me it would be absolutely clear that both involved “captures” are defined in the same way when they are used in a combined situation. So, “capture” as the base idea for observation in Isardam would have to be understood as a Marscirce type capture in this combination. Do we agree insofar, or is this already the splitting?

Let me adjust my simple example from before, considering your remarks: Marscirce+Isardam, wQa3, bQg1, bRf3; 1.wQa3(via d1)xbRf3 = legal? In fact under Marscirce (solely) the wQ could capture bRf3 from EVERYWHERE on the board (via d1), in spatial terms as long as d1 and e2 are free (or the wQ already may even occupy d1), because the Q can move / capture d1-e2-f3. Do we still agree?

The only open question (in my opinion), is whether an additional Isardam restriction has to be respected here by a bQ not only observing the capture square f3 BUT ALSO observing the (interim) starting square of the capture move d1? AND from which square an observing bQg1 is acting – from her rebirth square d8 (because of the Marscirce condition!!!???), or from her current position in the diagram!? AND conversely, whether the wQ would also observe the bQg1 from her (interim) starting square d1 for the Marscirce capture move of bRf3!? Because obviously “at the end of this capture move” there is no observation at all between both Queens on f3 and g1, but what about the “interim” positions?! You write: “A piece observes another one from the starting point of its capturing scope…”; would you mean the Marscirce capturing move wQa3xbRf3 is Marscirce+Isardam ILLEGAL, because of wQa3 observes(!) bQg1 (via d1), and at the same time bQg1 observes(!) wQd1 (via d8)…?

Actually that’s all I want to find out – is there only ONE unique answer or can there be different valid(!) interpretations for the observed / observing squares on which opponent pieces of the same kind are (Marscirce/interim?) placed … And I honestly do apologize if I misunderstand your comments…

Kevin Begley
April 22, 2023 00:24

In Mars Circe + Isardam, the position was illegal before the move.
First test: if black were on the move, could bQ at g1 capture wQ at a3 (under the list of all conditions, minus Isardam)?
If yes, the position is illegal (by Isardam).
Can that happen? No. We passed that check.
Second check: if white were on the move, could wQ at a3 capture bQ at d1 (under the list of all conditions, minus Isardam)?
If yes, the position is illegal (by Isardam).
Can that happen? Yes, it can. In Mars Circe, a wQ at a3 can certainly capture a bQ at g1 (place the wQ from a3 onto the empty d1 square, and you’ll clearly see what captures it can make — clearly it can capture g1).
Thus, the position you inquire about was illegal from the outset (you can not ask whether a move is legal from a position which is not legal) — this is not a valid question. Try asking your question again (this time from a legal position).

Kevin Begley
April 22, 2023 00:50

Err, I meant to say:
Second check: if white were on the move, could wQ at a3 capture bQ at g1 — I meant to say g1, not d1! — (under the list of all conditions, minus Isardam)?
If two like units are on the board, and EITHER of these units could capture the other, in the given position, under the rules of play (all conditions minus Isardam), then the diagram is illegal.
So, your question is like asking: “in an orthodox chess game, with wQd1, bRf3, bKg1, wKa8 can white legally play 1.Qxf3?”
First, white can not legally be on the move in the diagram.
OK, theoretically, in the interest of learning what point you are attempting to make here, yes, this would be a legal move for white, if the diagram were legal. It’s true, many fairy problems do begin from a position which is not technically legal (according the rules), so, I’ll give you some latitude to make your point.
Please understand that it’s forgivable to begin with a position with fewer than 32 total units in PWC (or to begin from a retroactively illegal position in monochrome Chess); it is less forgivable to begin with a position which is already in stark violation of the condition (e.g., both players are in check), and ask what moves are theoretically legal.

Walter Lindenthal
April 22, 2023 12:42

MANY THANKS, I hoped for such an assertive reply! It may sound strange, but I asked the “wrong” question intentionally, because I couldn’t relate any more some of what has been written before, so this was intended as some kind of challenge! I fully agree, the position was illegal even before the move – and so is any position of the wQ on the board here in spatial terms (because of Marscirce!; and not only those few “restricted” situations based on “scopes”(???) as presented on April 18, referring to (quote) “pieces blocking each other“, or “move everywhere on the a-file or 3rd rank, but capture only on d4” – notions that I can’t understand, but may be I misunderstood them)!!!

Now lets introduce an additional bSe1 to my simple example from before: I assume this will result in a legal capture wQa3(via d1)xbRf3. Or is there still an observation of the wQ’s advance rebirth square d1 “during” the capture move by bQg1 via(!!!) her own advance rebirth square d8 (because of the “emancipated” Marscirce condition!; it was also written (quote): “Assuming that observation is based on the possibility to capture, the capturing scope must be recognized before any observation check, which consequently obtains the status of a 2nd level consideration“)? And the wording of one of the rules lately was (quote): “A piece observes another one from the starting point of its capturing scope, and you must not position any piece in the capturing scope of a similar piece in Isardam” – attention to the word “scope”, it’s not referred to as “square” (where the piece actually has to be positioned)!!!??? The word “scope” has a multitude of possible meanings in dimensional terms… And also one has to interpret the meaning of “…to position any piece in the capturing scope…”, provided a capture move under the given regime only starts from the advance-rebirth square, is the piece positioned there??? – (quote): “The Q IS already on d1 when “leaving” for a capture…”

Important note! I do NOT want to pit comments against comments or so, but that’s all I’ve got to interpret. I just wanted to embrace the opportunity to learn from the best! Unfortunately I couldn’t understand fully… Sorry!

PS: I also DO appreciate your reference to “illegal fairy diagrams” as starting positions…

Kevin Begley
April 22, 2023 14:10

If you wish to learn from the best, I’ll not delay you further from your goal, except to say that, yes, obviously, adding bSe1 would make Qxf3 a legal move (ignoring, of course, that there are no Kings on the board).
So, if you have something to illustrate here (which I must assume, given your slow unfurling of this very simple example), this would be a fine time to come to your point.
Otherwise, I hope I’ve helped.
ps: don’t get too hung up on understanding scope. If you know the fundamental definitions, your Jedi training is complete.

Walter Lindenthal
April 22, 2023 17:19

Ok, no observation between both (interim) advance-rebirth squares d1 and d8.
THANKS! The Force restocks…

Manfred Rittirsch
May 3, 2023 12:50

I am sorry that I failed to find the right words to make you understand my concept of capture & observation. Maybe I can express it better in German:
1) Eine Figur verfügt über einen Zugangsbereich gemäß ihres Typs.
2) Der Zugangsbereich einer orthodoxen Figur mit Ausnahme des Bauern ist identisch für schlagfreie und schlagende Züge und geht vom aktuellen Standfeld auf dem Schachbrett aus, der durch das Figurensymbol markiert ist.
3) Es gibt Figuren, die unterschiedliche Zugangsbereiche für schlagende bzw. schlagfreie Züge haben (= „Schützen“), zu welchen auch der orthodoxe Bauer zu zählen ist.
4) Es gibt Märchenbedingungen, die den Figuren unterschiedliche Zugangsbereiche für schlagende bzw. schlagfreie Züge auferlegen.
So kann z. B. bei der Märchenbedingung „Anticirce“ der Zugangsbereich für schlagende Züge kein einziges Feld enthalten, weil das zugehörige Wiederentstehungsfeld besetzt ist.
5) Befindet sich ein Feld nicht im Zugangsbereich eines Steines A für schlagende Züge, sondern allein in jenem für nichtschlagende Züge, so wirkt die Masse eines dort befindlichen Steines B unabhängig von seiner Farbe blockend und sperrend für Bewegungen des Steines A.
6) Für die Märchenbedingung „Marscirce“ besteht der Unterschied für schlagende bzw. schlagfreie Züge allein im abweichenden Ausgangsfeld, wodurch geänderte Zugangsbereiche entstehen.
Genau dann, wenn man die Absicht hegt, einen Schlagzug auszuführen, stellt man sich einfach eine Situation vor, die bezüglich des zur Ausführung des Schlages bestimmten Steines, soweit sich dieser nicht gerade genau auf seinem Ursprungsfeld befindet, nicht jener im Diagramm wiedergegebenen entspricht, sondern a) keinerlei Masse auf dem im Diagramm markierten Standfeld jenes Steines und b) diesen Stein auf seinem Ursprungsfeld aufweist. Dabei gibt es weder zeitlich noch räumlich ein „interim“, die Umschaltung findet allein in der Vorstellung statt.
7) Ein Stein beobachtet einen anderen Stein genau dann, wenn dieser sich in seinem Zugangsbereich für schlagende Züge befindet. Die Partei eines Steines muß am Zug sein, um schlagen (und damit einen gegnerischen Stein beobachten) zu können.
Aus 6) & 7) folgt, daß der Zugangsbereich eines Steines A für schlagende Züge inklusive des entsprechenden Ausgangsfeldes für Überlegungen bezüglich der Beobachtungswirkung eines Steines B irrelevant ist, weil bei derartigen Überlegungen immer davon ausgegangen wird, daß Stein B „am Zug“ ist, so daß A eine passive Rolle zukommt. Für Überlegungen zur Beobachtungswirkung des Steines A verhält es sich genau umgekehrt.
I hope these explanations were appropriate to clarify my point of view.

Manfred Rittirsch
May 3, 2023 12:54

Bei Punkt 2) ist der Vollständigkeit halber hinzuzufügen, daß die Zugangsbereiche eines Königs im speziellen Fall einer Rochade, die ja nur schlagfrei erfolgen kann, ebenfalls unterschiedlich sind.

Walter Lindenthal
May 4, 2023 12:15

Vielen Dank für Ihre besonderen Bemühungen! Now I can see what you meant with “spatial relationship” and “scope”. Your bullets 1-6 are self-evident and undisputed (and of course, a “specific” line can be blocked). You formulated them in a very formalized style that I misinterpreted / misunderstood at first – my way to see them is more easygoing. Even “interim” (or “via…”) seems no issue at all, it’s just wording. I do agree, the positional “starting square” is of no importance here (we need not discuss the issue of a position that is illegal before even moving), only the advance-rebirth square (and of course the capture/target square) are relevant! Also, the capture (with an empty rebirth square e.g. Anticirce rebirth [unless e.g. Anticirce Assassin, etc], or an empty advance-rebirth square [unless occupied by the capturer itself, or in case of Jupiterparacirce!]) must be feasible…

Your bullet point 7 requires more of a consideration. You presuppose that only the side which is on the turn to move can capture, consequently only this very same side can establish observation (because observation is based on the idea of capture; quote: “Die Partei eines Steines muß am Zug sein, um schlagen (und damit einen gegnerischen Stein beobachten) zu können”). This is true if you set the higher priority to capture! And you are right, this avoids some problems here! But is this the only “thinkable” way? – and THIS was exactly my question here…

I think it’s always best to present a simple example, e.g.: Isardam+Marscirce, wQb2, nPd4, bQg7. NO move of nPd4 would create observation here (not even if it were a neutral Berolinapawn, opening the d-line – and this is in stark contrast to a situation without the Marscirce condition)! However, 1.wQb2(via d1)×nPd4, this capture move still does NOT create “active” observation from the white side’s move (on its own terms), nevertheless it “results” in an illegal observation because of the black Q’s following turn to move (the bQg7 could then capture the wQd4 via d8 = observation Marscirce-style)! Now, if it were a white series-mover (except the last move in the series) it should be legal, because it is not the black side’s following turn to move – at least this should be the consequence of your interpretation. Thus, is observation always depending on the other side’s turn to move…? May I refer to the definition in Märchenschachlexikon: “Serienzüge: … Illegale Selbstschachs sind währenddessen untersagt, ein Schach ist nur mit dem Abschlusszug der Serie möglich. Dies gilt entsprechend auch für Rückzüge: Hier kann nur mit dem letzten Zug einer Rücknahmeserie in ein Schach zurück gezogen werden. Bei einer solchen Rücknahmeserie muss nach jedem zurückgenommenen Zug eine legale Stellung vorliegen.”… the last sentence translates as “… after each (retracted) move a legal position is required”. Would the requirement “legal” only be restricted to check here?

Now let’s change nPd4>f3 and 1.wQb2(via d1)×nPf3: Based on your interpretation (and the discussion) it was shown there is NO observation by the bQg7(via d8!) of the (“interim”) square d1, because “during” executing the white move it is not the bQ’s turn to move/observe, and once the white capture move has been finished there is no observation (via d8) left by the bQ. So, this capture move should be legal in this interpretation… But could the judgement be different if you assign the higher priority to observation, an observation that is defined by the Marscirce capture style? Would thus the bQg7 (“passively observing” via d8) observe the wQ “starting” (= via…) her capture move on square d1…? I do not know…

I feel this “restrictive” discussion is unique for the advance-rebirth situation when it combines with a condition where both of them are based on capture (observation). In all the other situations with after-capture rebirth situations (e.g. Anticirce+Isardam, or even take&make+(Anti)Circe+Madrasi, etc) this specific problem doesn’t exist at all, because here the capture “must” have happened first, creating (positional) “facts” instead of (virtual) “assumptions”. So, then you could easily distinguish between different “priorities”, like e.g. the observation emerges immediately after capture or only after rebirth (or even before/after the make part-move – see the different implementations between Popeye and Winchloe)… The importance of clarifying these “structural” sequences seems to get even more important once you exchange Marscirce with other Versions of an advance-rebirth style, like Jupiter(para)circe or even their Cuckoo-Variants…

Manfred Rittirsch
May 5, 2023 14:44

Thank you for your further analysis! I am afraid that a formalized way of expressing things is indispensable when it comes to rules, definitions etc. But I now realize that the 2nd sentence of my bullet point 7 is simply wrong! Please excuse my negligence! What I wanted to express is that
a) (capabilities of) 2 pieces must always be examined separately
and
b) when examining piece A, this piece – during the process of examination – is assumed to be the active one (“on the move”)
and
c) as a consequence, (the virtual 2nd instance of) a passive piece B – regarding the current process of examination – does not exist on its starting square for captures, if this is different from the diagram square.
And of course you must always examine all pieces on the board regarding observation (one after the other).

Walter Lindenthal
May 5, 2023 18:14

ok, this clarifies your opinion on the combination Marscirce+Isardam for me (quote: “during the process of examination“). So, you have to “finish” a single one “examination (of conditions)”-step first before you “start” the next one, they must not be intertwined! Concerning my example: “finish” examination of legality of the white move wQb2(via d1)×nPf3, before you start to examine legality of a (virtual, if at all existing) black capture eventuality bQg7(via d8)×(?)wQd1/f3. After the first examination step the wQ already arrived at f3, so the following examination step cannot come across her on d1 any more, and f3 is out of reach of bQ… thus, legal!

Isardam is about legality (thus examination!), in contrast to Madrasi which is “only” about (even voluntarily entering!) paralysis. Presumably this fact establishes the difference between AntiCirce+Madrasi (with priority options: paralysis already on capture square or only on rebirth square, if bQg4 with/without bPf4: 1.wQa3×bPb4 [wQd1??]) versus Anticirce+Isardam (with NO options, because it wouldn’t matter “when” illegality occurs during the move, the move IS illegal!)…

Manfred Rittirsch
May 7, 2023 20:16

If I understood your “priority option” regarding AntiCirce+Madrasi properly, you are suggesting that a capture of piece B by piece A may be considered illegal due to a paralysis on the capturing square by a piece C stopping solely the required transfer to the rebirth square of A. Note that for this scenario you would need to deviate from the standard concept of a “move” by splitting it (the “activity” of piece A) in 2 parts (applying a real chronological “interim” state) in order to examine a capability of previously inactive piece C.

Walter Lindenthal
May 8, 2023 12:27

Yes, indeed, it’s about splitting a move into 2 (or more) logical phases – think e.g. of Andernachschach+Anticirce, or Masand+Isardam, or take&make+Circe+Paracirce, etc… or some problems of prioritization when capturer and captured piece have the same rebirth square…! And maybe think of part-moves like take&make and the discussion on Circe+take&make versus take&make+Circe. Is there really a “standard concept of a move”, or is there only some kind of tradition based on the fact that conditions were more or less orthodox-oriented in the old days before Circe and take&make… (may I refer to the most interesting article of Vlaicu Crişan in feenschach 220/222/226 “The unbearable easyness of the fairy composition“)? Whenever a “single move” is divided into logically separable phases and at the same time “combined conditions” have to be applied, it simply seems to be a matter of priority which condition has to be considered overridingly – it’s not always possible to apply several conditions truly simultaneously and with equal value, prioritization is a matter of necessity (or just resulting from some “natural primacy”, possibly based on habit). So, here e.g. paralysis could emerge for a capturer at the capture square dropping out the necessity to complete the capture move with rebirth (= “priority Madrasi”), or alternatively you also could stipulate it the other way around so that rebirth has to happen first, only then you examine observation (= “priority Anticirce”). As far as I know Winchloe and Popeye handle this issue in some respect fixed but differently! And it is not about right or wrong, it is just about definition!

As I understand it, the answer to this whole issue is rather complex. On the one hand you have different definitions of “observation” (which establishes the base definition for many other conditions): Winchloe seems to require a “real practicability” of the capture move (else the capture cannot be truly executed, as a consequence of that –> no capture means no observation insofar). But Märchenschachlexikon has a less stringent definition (see before: thus, restrictive fairy conditions (e.g. an occupied rebirth square under Anticirce?) seem not to hinder a “prioritized” observation); at the same time this specific exceptional rule is not included e.g. in the separate definition of Superguards (resulting in different outcomes there)! So, is the rebirth of a capturer under Anticirce an inseparable part of a capture move or not? As far as a capturing King (Anticirce) is concerned there is no check threat on the capture square, only possible after his rebirth on the rebirth square, but all the other thinkable situations… where is a standard definition? And on the other hand there are many examples of variably applied priorities already, I had the honor to act as a TD for feenschach TT80 “priority setting with fairy conditions” (results to be announced soon in feenschach). Some older versions mostly trusted even in the sequence(!) of terms in the conditions statement in order to set the priorities (see e.g. the old issue of Monochrome and Circe)… All of that is the reason why it seems prudent to indicate the intended priority-setting for a presented problem whenever there might be some doubt about it (even if the composer him-/herself don’t have doubts, others might have them…); but of course not always there is doubt… And this should NOT AT ALL restrict creativity in any way, quite contrary it should expand the chances for creating even more interesting problems while clarifying the framework conditions…

And finally (as a rather descriptive example for this discussion) let me point to E. Huber yacpdb 362029 (h=2.5 Anticirce+Isardam): In both solutions the move 3.e×f1=… should create illegal observation in case of “priority Isardam”, however the move is legal in case of “priority Anticirce” (which might be applied as “natural primacy” here)… These are my thoughts, not sure about their accuracy! THANKS!

Walter Lindenthal
May 8, 2023 12:36

I forgot to mention explicitly: Rebirth is a condition, it is not a move, neither is it part of the capture “move” under Anticirce!

Georgy Evseev
May 4, 2023 21:26

I thought it over once more.
As Isardam definition based on Madrasi definition.
A number of fairy conditions allows for asymmetric Madrasi, when piece A paralyses B, but B does not paralyse A.
Here is a simple example:
White : Kd2 Re6 Be1 Sd1 Pg4f3 – Black : Kf4 Rc6 Be8 Pg5d4h4d3
#1 Duplex,
Here both rooks are paralysed and there is no solution.
Let us add black knight to h1. Now WinChloe seems that Rc6 is not paralysed and black can mate 1…Rc2#.
Let us add black knight to a8 instead. Now WinChloe seems that Re6 is not paralysed and white can mate 1.Re4#.
If we add two knights then both mates become available.
I need to add that Popeye never sees any mate here as allowed.
After very sketchy review of WinChloe database I have found a number of Madrasi/Isardam problems with such asymmetric paralysis, including one my problem (173233, though with different approach). All these problems are not Popeye compatible.
I understand that original Madrasi definition includes words “each other” and so does not include asymmetric cases. BTW, the FCCP definition is simply wrong as it does not mention that the units should be of same kind.
WinChloe approach seems to me more logical, and I think that situation may be corrected by applying the following definition (probably, it requires a lot of editing).
If piece X may be physically captured by opposite side with the piece of same kind (not taking into account turn of move, legality of position and legality of move), then piece X is paralysed and it can neither move, capture or check, but retains the power to paralyse other units.

shankar ram
May 5, 2023 15:16

Thank you for pointing that out, Georgy!
The correct definition for Madrasi: “a unit (not K) when observed by a similar enemy unit is paralysed, and can neither move, capture nor check, but may paralyse in turn” was there in the FCCP version dated 23-Mar-2021. Subsequently, it was revised and the word “similar” got left out. Apologies.

Last edited 1 year ago by shankar ram
Walter Lindenthal
May 5, 2023 17:34

Again this system qualified my reply as:
Walter Lindenthal
May 5, 2023 17:28