A great idea to use the familiar ring R-B-S for selfmating!

One can see that the realization was not easy...

Quoting Nikola Predrag:

But it's mysterious to me how easily the name Zilahi is given to a quite opposite essence.

Nikola, I composed this for SuperProblem TT-109. Here is an original definition of theme:

[qyote]" Typical selfmate Zilahi. In one variation (a) the White capture a black piece/pawn which mates in the other variation (b). In the variation b) the White capture the black piece/pawn which mates in a). Cyclic form of the theme is welcomed.".

That's why I named it Zilahi. Although I agree it's not the best name.

This problem did not make an award. Judge commented this problem with those words:

Quote:

"Too much repetition of moves and/or sequences of moves;".

In threat and two thematic variations there is no repetition of moves at all. There is only repetition in side variation 1...Kf4!.

It is often questionable to use a theme name from one problem form in a completely different form. It happens that just the superficial pattern is transferred, not the real meaning of the theme (example: Umnov in helpmates).

But in this case, there is some sense in using the Zilahi name. You could argue that WHITE sacrifices one black piece (that could have helped him reach his goal of being mated) in order to let Another black piece mate him.

It semms clear that it is not Black who sacrifices here; he has no choice!

Rodolfo, intention gives a true meaning of a theme and superficial appearance could give a false meaning. Scientific approach searches for a true meaning. False interpretation of the appearance is a tool of deceivers who try to sell nothing for something.
For the White's task, 1...e1R and 1...Sg3 are the same, both guarding e2&e4. It is one variation for White who doesn't have to care which move Black will choose, only what will be the effect.
You may see it as unpleasant but for a serious judgement, your personal taste should not have a significant effect on such a complex idea.

Kjell has pretty convincingly shown that actually there is the essence of White Zilahi but using the black pieces.

I'm sure that a theme like the " n defenses by different black pieces (with the same effects, please!) all followed by the same mate " could attract some problemist but not the best Nicola: the composer.

I am not attracted by 2 equivalent black moves followed by the same white continuation, and I never said so. One of these moves is superfluous for the content, that shows the imperfect economy of realization. Since White must fulfill the task and Black opposes, this has not a significance of a dual. It is a minor imperfection, present in many direct and selfmate problems. Also, 1st black moves are not directly thematic, so this imperfection could be neglected or at least treated as a detail.
I would say that it might decrease the value of the realization for few percents but not more than 10 percents. If the complexity of the idea is arbitrarily worth 20 points, the realization would be worth at least 18 points. This would still be more than a perfect realization of some simpler idea.
(But for instance, a possibility of mates by two black pieces in the same variation could not be thematic, since a mate by a particular piece makes the theme.)

One commentator argues that the two defenses ensemble can do part-time work: sometimes it is ok, but not always of course... Whatever the theme of what we're talking about, in my view the "two or more concurrent defenses by different pieces" is always a flaw (wich could be added or not to others). It is not part of a proper chess Language and, when unavoydable, could be accepted only in case of exceptional task (not the absolute masterpiece). Obviously when we are forced to evaluate any work on the basis of stylistic factors, we previously know most of the answer. The method of evaluation proposed by Nicola is too complex as it envolves the knowledge of arithmetic. There is already an example of the "here cyclic" theme without repetitions and side variations and with amazingly homogeneous play: cycle of capturing (W2) and unpinning moves (W3), etc.
But this will not discourage a great composer to engag? in ambitious projects.

I know that I may be slightly off-topic, but I have to grab the chance given by Kjell's comment #5 re. the use of the Umnov theme in helpmates. I fully agree this is a typical example of a theme which is paradoxical in directmates, but there is no meaning at all using it in helpmates (and surprisingly it is used even in the FIDE album for helpmates!) Most of the readers probably know that the term "Follow-My-Leader" (FML), coined by Chris Feather, is more appropriate in helpmates and it should be adopted in this genre.

You can see FML in the threemover 381 published here yesterday...

There are quite a few theme names freely transferred from one genre to another based on formal similarities, but often with questionable basis - besides Umnov in h# and Zilahi in s# also the following come to my mind: Dombrovskis in #3, Ceriani-Frolkin in non-retro, flight-giving key in r#. Obviously name is one thing and the real value something different.