Website founded by
Milan Velimirović
in 2006

9:44 UTC
ISC 2021



Remember me

Forgot your
Click here!
to create your account if you don't already have one.

Rating lists

MatPlus.Net Forum Fairies Circe Assassin
You can only view this page!
Page: [Previous] [Next] 1 2
(1) Posted by Paul Raican [Tuesday, Dec 1, 2009 23:50]

Circe Assassin

Recently, I published in StrateGems the following problem:
P. Raican - dedicated to Didier Innocenti
(= 6+2 )

Circe Assassin
Definition: Like Circe, but when the rebirth square is occupied, the occupant is « assassinated » (replaced) by the captured unit. A piece is, so, self-protected if it is on the original square. Another result: if the black King is, let say, in b7, then it is in check if a black Pawn on the b file is attacked.
Solution: 1...Kxc7(Ra1) 2.Rh1 Qxa7(Sg1) 3.Qc6+ Kd8+ 4.Kh8 Qe7 5.Rh7+! (black King is in d8, so attacking the black Queen means check) 5...Qf8#
My question is: who invented this magnificent fairy condition and when?
(Read Only)pid=4392
(2) Posted by Siegfried Hornecker [Wednesday, Dec 2, 2009 01:43]; edited by Siegfried Hornecker [09-12-02]

Does "Assassin only" already exist, where one can only capture if he can assassinate a piece?

(= 8+5 )

SH, Original
Assassin only

1.Ke7+ Rf1! 2.Qc8! stalemate (2.Qa8? Rxb1!, 2.Kc8? Rd1!)

Not sure about 2.Re8, though.
(Read Only)pid=4393
(3) Posted by Joost de Heer [Wednesday, Dec 2, 2009 19:20]

I think it was invented by Didier Innocenti. I can't find (nor remember) anymore where I got this information though.
(Read Only)pid=4395
(4) Posted by Paul Raican [Saturday, Dec 22, 2012 20:25]

Now, we know the answer. This genre was invented in 1978 by Romeo Bedoni (see
(Read Only)pid=9339
(5) Posted by Neal Turner [Monday, Dec 24, 2012 13:13]; edited by Neal Turner [12-12-24]

I don't want to spoil the fun, but...
It seems there's many fairy conditions that rely for their effects on the idea that the King can be captured.
The fact is that in Chess the King is never captured.
This is stated both implicitly - pinned pieces can give check - and explicitly - the FIDE rule that in Blitz capturing the King is an illegal move leading loss of the game.
So in the first example it would more consistent if the effect of the K sitting on d8 would be simply to make the bQ immune from capture - with no 'check' occuring when the Q is threatened.
(Read Only)pid=9345
(6) Posted by Jacques Rotenberg [Monday, Dec 24, 2012 13:36]

What really disturbs you ?
It is common place in fairy chess to break any rule of the FIDE.
Here a new kind of capture is defined, that's all.

Your proposal seems fine, but you should find a name for it.
Perhaps "Circe assassin Loyaliste" ? Loyalist because every piece but the King can be assassinated...
(Read Only)pid=9346
(7) Posted by Nikola Predrag [Monday, Dec 24, 2012 16:20]

It is not the same capturing a King and checking it. If a hypothetical capture is not legal King is simply not in check. If a capture of a King does't happen in the play, what is then against the orthodox logic. In a chessgame bK is initially not in check, obviously, but that is exactly because for instance 1.Sg1xe8 is not a legal move.

Sigfried, I still don't see why bR can't play to h1.
(Read Only)pid=9347
(8) Posted by Kevin Begley [Monday, Dec 24, 2012 18:33]; edited by Kevin Begley [12-12-24]


>The fact is that in Chess the King is never captured.

I've gone back and forth on this issue.

But, before I come to that, let's clear the air about some terms being tossed around.

1) Annihilation consistency is maintained.
Actually, only the threat to annihilate a King constitutes check (capture of a King is no more than a dance step, which cannot be played in orthodox form -- only because capture there must always result in annihilation).
Circe Assassin maintains a completely consistent "fairy orthodoxy," when it comes to annihilation of the King.
Check is defined by a threat to annihilate a King, and you need not ever contend with any actual instance where this threat is carried out (captures of Kings yes, but not annihilation).

2) Capture consistency requires some clarification.
The description for this condition seems to falsely imply that the captured unit, upon rebirth, may capture (and annihilate) the occupant of its home-square.
This is a poor description, which highly muddies the understanding.

The reborn unit, under the modified rebirth rules in Circe Assassin, is NOT responsible for capturing, nor assassinating, nor annihilating its home-square occupant -- not by ANY standard definition of these terms!
It would not be a consistent form of the term capture, were a reborn white unit to do this to the same-colored occupant (royal or otherwise) occupying its home-square.
So what really is going on here?
Black has simply annihilated the occupant, standing upon the captured white unit's home-square, via capture and rebirth of the captured white unit.
All this is made possible by modification to the rebirth rules -- and that's all that occurs here, nothing more, nothing less!

Circe Assassin maintains a completely consistent "fairy orthodoxy," when it comes to both capture, and annihilation of units (royal and otherwise).
There is no violation -- only some misunderstanding, caused by a poor description.

However -- and here may be where Neal failed to make his key point -- Circe Assassin is not consistent with the Circe Family.

3) Simplicity and Circe consistency.
From the perspective of rule-set simplicity, we would prefer to avoid special case rules.
So, at first it seems the Circe Assassin inventor chose well -- additional rules to prevent royal captures, and prevent rebirth onto royal occupants, would be just that: additional!

But, when we consider what Neal's suggestion -- an additional rule to further modify rebirth, in cases of Royal Occupancy -- should actually be called -- "Circe Assassin (Rex Exclusive)" is a much better label for your interpretation, than the one Jacques suggested! --, we begin to see that Neal has a valid point (though perhaps not well articulated).

The consistent Circe convention is to make "Rex Exclusive" the default position (and offer "Rex Inclusive" as an option).

At first, I wanted to argue against Circe consistency, for two key reasons:
1) Where possible, we should preserve the inventor's intent as the default form, and
2) The inventor chose the simplest form of description.

But, there is no good argument for preserving this inventor's intent, when it fails to maintain the consistent intent of the higher form (the entire Circe family).
If there is to be consistency, Circe Assassin must first maintain allegiance to Circe!

And, if we are going to argue simplicity -- and it is quite correct to view Rex Inclusive as the simplified form -- we must recognize that simplicity arguments can be made for virtually all Circe instances!
Indeed, Circe itself would be simplified by defaulting to the "Rex Inclusive" case.

Special case rules to describe royal capture/rebirth/occupancy consistently prove more complex.

Personally, I would prefer to modify all Circe forms to default to that simpler form ("Rex Inclusive"), and always offer the option of "Rex Exclusive."
The simpler our default position, the better -- not only to describe, but also to simplify the process of progressive development.

Popularity should never dictate these matters; though, I often wonder whether chess problems were invented by people seeking to create some new form of popularity (it is the focus in so many of our contests).

So, the form of Circe Assassin has pros and cons:
It should be consistent with the Circe family form (which defaults to "Rex Exclusive").
But, the Circe family should have defaulted to "Rex Inclusive."
(Read Only)pid=9349
(9) Posted by Ian Shanahan [Monday, Dec 24, 2012 20:51]

Well said Kevin. We already have Circe RI (formerly known as "King Circe", according to an article by Norman Macleod in "Chessics").
(Read Only)pid=9350
(10) Posted by Kevin Begley [Tuesday, Dec 25, 2012 00:00]; edited by Kevin Begley [12-12-25]

Thanks Ian.

Though, I'm not sure that we are on the same page (entirely)...
I am already well aware that Circe forms do provide a "Rex Inclusive" option.

As I stated, Neal has a legitimate point, if his argument is that Circe Assassin should have followed the established convention, by defaulting to "Rex Exclusive," and offering a "Rex Inclusive" option.

We could make similar "convention consistency" arguments about a vast number of circe (and anti-circe) forms.
This includes many (like PWC, Take&Make, etc) where the inventor not only discards convention, but also fails to recognize the anticipation of their form (the two mentioned are a form of circe, and of anti-circe, respectively).

In the interest of consistency, simplicity, and continued logical development (read: progress), I would prefer to see the delegates (counseled by a variety of fairy experts: inventors, programmers, composers, variant players, etc) exercise their neglected authority to standardize, codify and sanction fairy element inventions.

I reject the argument that the inventor can ignore anticipation, ignore naming convention, and ignore the need to maintain consistency with the default rules of the form.
But, I also reject the argument that the inventor sets an unalterable precedent.

The foundation of fairy chess elements must be a simple, logical, consistent, and codified set of rules.
This can not be dictated by popularity, nor by inventor preference, nor even by publication priority.

Circe is MUCH bigger than the default rules of its original invented instance -- it constitutes a form, under which a large family of conditions now reside.
The most concise definition for the larger form (something few inventors appreciate from the start) must dictate the default rules for all sub-instances (including the original instance).

I consider "Rex Inclusive" to offer the most concise definition for the larger circe form.
Whereas the original inventor deserves credit for having invented the form, this does not necessitate that the original instance match the evolved form!
Logic dictates that we distinguish between the original instance, and the evolved form; but, the larger community tends to ignore this!

Sub-inventions should be named & classified according to the priority established by the evolved form.
For example, the inventor of Platzwechsel Circe can not ignore that rebirth already fell under an anticipated form (Circe) -- we can not accept the author's original naming convention ("PWC" is a name which should be rejected), nor can we accept their new assertion of default rules (pawns reborn onto the 1st rank should default to those of the larger form)!
Same goes for Take&Make (which constitutes a poorly named, anti-circe form, with an inconsistent set of default rules).

It doesn't matter whether the majority of these problems need to assert an additional condition (to over-ride the concise rules to which the form defaults).
This is not a popularity contest -- our highest priority must be to codify a clear and consistent form!

I believe that "Rex Inclusive" constitutes the most concise definition of the circe form (just consider how it better applies to the anti-circe form).
Therefore, even though I am well aware that "RI" is less popular (and, there may be good reasons to consider it a less interesting default), "Rex Inclusive" should be the default rule (found in the default instance of both Circe, and anti-circe) -- even when it may disagree with the original instance.

This should lead to no confusion, providing that invention credit begins by describing the form, and then considers the instance of it.
You can not say that Petko Petkov invented neutral Kings (if memory serves, this is true), without first explaining that T.R.Dawson invented the neutral form!

ps: if you want to investigate the "rex inclusive" default further, just consider how it applies to the anti-circe form!

At present, Anti-Circe essentially describes an anti-form (capturing unit is reborn, according to the circe instance, and the captured unit is annihilated) of a non-default circe.
Anti-Circe changes two default rules:
1) "Rex Incluslive" (rebirth also applies to Kings), and
2) "Strict" (the move is not legal, unless rebirth occurs).

And, to make matters worse, we don't have any default form of anti-circe.
We have to types (this is very bad form)!
No wonder I can seldom remember which is which.

Whereas the "Strict" option is more complex (than its default alternative), the "Rex Inclusive" option is actually simpler than the alternative (requiring additional explanation).
It is easy to see which better serves as the default.
(Read Only)pid=9353
(11) Posted by Jacques Rotenberg [Tuesday, Dec 25, 2012 04:49]; edited by Jacques Rotenberg [12-12-25]

Sorry Kevin not to agree with you on several points

1) Take & Make is not connected with circe and and not with anticirce.

2) "Circe Assassin (Rex Exclusive)" is certainly not a good name
-----> a) because it is unclear : what is the king excluded from ? capturing ? being captured ? no - only being "assassinated" that means being captured in a specific way
-----> b) it does not say the original solution Neal proposes : he says the capture will be forbidden. OK why not ? but it is far from being the simpliest way or even a "natural" way.
you could imagine someting else a) in that case the capture will be orthodox, b) the re-birth will take place on another square, c) the re-birth will take place when the king will leave d) anything else - there are many options!
-----> c) the circe is already rex exclusive so that a second time rex exclusive is somewhat confusing.
(btw I am not defending the proposal "loyalist" I made)

3) you say :
'...The foundation of fairy chess elements must be a simple, logical, consistent, and codified set of rules.
This can not be dictated by popularity, nor by inventor preference, nor even by publication priority...."
This is most criticizable. I would say on the contrary : popularity, inventor preference, ordinary practice, are stronger than any other reason.

4) you speak of anti-circe - which is not really connected with the question of Neal - without underlining how "bad" a name it is.

5) 6) 7)... many other points...
(Read Only)pid=9354
(12) Posted by Ian Shanahan [Tuesday, Dec 25, 2012 05:18]

@Jacques. The terminology "Rex Inclusive" [RI] and "Rex Exclusive" [RE] are already well-established and -understood.

@Kevin. I agree with your point about RI being the default form. (Incidentally Bastian de Haas and I independently invented "Strict Circe" during the early 1980s; I came up with the term "strict"!)
(Read Only)pid=9355
(13) Posted by Jacques Rotenberg [Tuesday, Dec 25, 2012 09:13]

That's why it is not good here. I agree.
(Read Only)pid=9357
(14) Posted by Neal Turner [Tuesday, Dec 25, 2012 11:39]

@Jacques - There's no need to apologise for disagreeing with Kevin! :)

Another (non-Circe) example: Isardam
Here we can defend a check by observing the King with the same type of piece as the one giving check.
The logic being that now the K cannot be captured because under Isardam rules it would be illegal for the checking piece to stand on the square.
But the K isn't being threatened with capture - it's just in check!

And don't get me going on the practice of naming anti-forms with the reversed lettering of the original.
(Read Only)pid=9359
(15) Posted by Nikola Predrag [Tuesday, Dec 25, 2012 15:00]

Discussion without clear definitions may be entertaining, but hardly leading to reasonable conclusions.
What are the exact orthodox rules about capture-threat and check and what possibly makes the crucial difference?
The mentioned Blitz rule is not a genuine chess rule, it absurdly deals with already violeted rule of parrying the check.
When the difference between orthodox capture-threat and check is clearly defined, it may be compared with fairy rules.
So far, it has been entertaining.
(Read Only)pid=9361
(16) Posted by Neal Turner [Tuesday, Dec 25, 2012 15:32]; edited by Neal Turner [12-12-25]

FIDE Laws Of Chess (
The objective of each player is to place the opponent’s king ‘under attack’ in such a way that the opponent has no legal move. The player who achieves this goal is said to have ‘checkmated’ the opponent’s king and to have won the game. Leaving one’s own king under attack, exposing one’s own king to attack and also ’capturing’ the opponent’s king are not allowed. The opponent whose king has been checkmated has lost the game.

So there's no such thing as 'orthodox capture-threat' which is the point I've been trying to make.
Of course in Fairy Chess there's nothing to stop us from making our own rules, but these must be made explicit.
(Read Only)pid=9362
(17) Posted by Nikola Predrag [Tuesday, Dec 25, 2012 17:21]; edited by Nikola Predrag [12-12-25]

I was talking about CLEAR definitions (definite as the word suggests).
"attack" is not yet defined in your post. The formulation of 1.2 contains a potentially missleading nonsense (no surprise :)):
>Leaving one’s own king under attack, exposing one’s own king to attack AND ALSO 'CAPTURING' THE OPPONENT'S KING are not allowed<
The first part of the rule makes the last part completely meaningless.

Attack is defined in 3.1:
>...A piece is said to attack an opponent’s piece if the piece could make a capture on that square according to the Articles 3.2 to 3.8.
A piece is considered to attack a square, even if such a piece is constrained from moving to that square because it would then leave or place the king of its own colour under attack.<

So, there is such a thing as 'orthodox capture-threat' after all, if you don't insist on the wording.

If >...such a piece is constrained from moving to that square..< because (shortly said) of a 'selfcheck' alone, the square would still be attacked and the opponents King in check. But if such piece (irrelvantly to own King) can't legally move to particular square, then it does not attack that square and therefore it does not check the opponent's King standing on that square!
(Read Only)pid=9364
(18) Posted by Kevin Begley [Tuesday, Dec 25, 2012 18:23]; edited by Kevin Begley [12-12-25]


Thanks for the information on the inventors of the "strict" doctrine.
I am always interested to learn more about who invented various fairy elements and forms.
Can you please provide a date as to when this was invented?

Also, I'd be interested to hear your views on whether the "strict" enforcement upon captures should be the default in the anticirce form (or the circe form).
Why don't they both default to the non-strict form, and provide strict as an option?

The same question for the RE/RI options....
No need for two options -- why not make one the consistent default, and simplify to a single option?


I agree with Neal -- there's no need to be sorry for disagreeing with me. :)
Some of what you said, in fact, has convinced me to rethink my position.

For example, "Circe Assassin RI/RE" might not exactly fit.
It is only useful to describe the ability to capture a King, and you are quite right to note that it fails to describe annihilation of a King by rebirth priority.
I've already agreed that rebirth priority does not constitute a "capture" (in any traditional sense of the word).
So, perhaps "RI/RE" are insufficient to fully describe all possible options of rebirth.

However, with respect to Take&Make, you haven't provided any reasonable support to your claim that T&M is not an instance of the Circe/Anti-Circe family.
It certainly does fall under the anti-circe form.

What is the anti-circe form?
1) The capturing unit is reborn, according to the specific instance (which govern both circe and anti-circe rebirth). Default rebirth is onto home-squares (same as the circe default).
2) Presently, anti-circe defaults to the "strict" form, which governs captures (captures are illegal, unless the legal rebirth follows).

What is Take & Make?
1) Captured units in this condition are reborn (according to Hartmut's new rebirth paradigm).
2) Captures are illegal, unless the legal rebirth follows.
3) Plus at least TWO non-default rules (both of which are rather poor):
a) Castling is now illegal with reborn units, and
b) Pawns are prevented from being reborn onto their 1st rank (more on this similarly bad idea later).

There is no denying that T&M constitutes a new instance of an anti-circe form.
That is not to say it wasn't a good invention -- I think T&M was a very good invention (and there are a number of excellent compositions which verify this).

But, the inventor failed to appreciate that his new fairy condition is nothing more than a new rebirth paradigm, for what is an anti-circe form.

Had this been realized sooner:
1) The inconsistency in naming convention might have been avoided.
Hartmut's invention denies the tribute owed to the inventor of the anti-circe form (circe form too)!
We expect tribute be paid to anticipatory problems -- the same principle should hold true for inventors of fairy forms.
2) We could have avoided the rather absurd default rule (preventing pawn rebirth onto the 1st rank).
Anyone who considers this a good idea, should have introduced it as an option spanning all circe/acirce forms.
If the inventor had been aware that existing anticirce defaults cover these situations (in a better way!), I very much doubt this alternative would have been introduced.
I don't dispute that it might present an interesting alternative to the Circe/Acirce default; but, I do not consider it to be in the spirit of these forms, to outlaw only a single type of retro-illegal situation.
Partial compliance with the orthodox retro characteristics of pawns, doesn't fully achieve the spirit of compliance that the inventor seems to have been striving towards (in fact, I believe a fully ortho-retro compliant option already does exist)!
3) He would have also realized that his invention has an anti-form (which I call "Circe T&M") -- where the captured unit is reborn, according to the movement pattern of the capturing unit.

As for some of your other points, I have more to say...
But, so as not to sidetrack the discussion, I thought it best to limit this Take&Make issue to a single post (as I am highly confident the matter can be resolved cleanly in this aside).
If you (or indeed anyone here) can provide ANY logical reason that we should not accept T&M as a member of the Anti-Circe family, I would love to hear it.
(Read Only)pid=9365
(19) Posted by Neal Turner [Tuesday, Dec 25, 2012 19:39]; edited by Neal Turner [12-12-25]

'I agree with Neal..'
Oh no!!!

But with regard to Circe Assassin - you can call it 'assassination', you can call it 'annihilation', but when one piece lands on a square occupied by another and the second piece is removed from the board, then I call that 'capture'.

Notice in 3.1 that they differentiate between attacking a piece and attacking a square:
Now of course 'capturing the opponent’s king is not allowed' is inconsistent with:

'A piece is said to attack an opponent’s PIECE if the piece could make a capture on that square according to the Articles 3.2 to 3.8.'

but it's not inconsistent with:

'A piece is considered to attack a SQUARE, even if such a piece is constrained from moving to that square because it would then leave or place the king of its own colour under attack.'

So the problem with 1.2 is that they're talking about the King (as a piece) being attacked, when to be consistent with 3.1 they should be talking about the SQUARE that the king is occupying.
(Read Only)pid=9366
(20) Posted by Nikola Predrag [Tuesday, Dec 25, 2012 20:30]

Kevin, your posts are not always entertaining. Reading them, I find myself very soon wondering "What was the initial topic". Without a clear critical point towards which a discussion leads, I simply cease to read. Perhaps later some critical point is reached, but I miss it. First thing is to agree what is Anticirce and what is T&M. Perhaps it was well hidden among hundreds of words.

Anticirce - capturing piece is reborn on a square determined only by the nature of that piece and the square of the capture, and rebirth square must be empty (2 ways of defining empty).

T&M - the set of squares on which a capturing piece may end the move are determined by the nature of the captured piece (this might partially remind on limited Supercirce) and on the open lines (this does not remind on Super, Anti or just Circe). The piece is not reborn, after visiting the square of capture it must move across the board TOWARDS some square. It is not reborn from "outside of the board" ON some square.
Well, orthodox chess is also a type of Anti-circe, a capturing piece is reborn on the square of capture.
(Read Only)pid=9367

Page: [Previous] [Next] 1 2

MatPlus.Net Forum Fairies Circe Assassin