THE LOGIC AND BEAUTY OF THREAT CORRECTION

by Marjan Kovačević, in collaboration with David Shire and Valery Shanshin

Part III

Some thematic fields have been universally defined, with certain criteria accepted by the majority of chess composers. However, in the area of correction play there are relatively arbitrary standards dependent on different traditions and times. For example, the basic concept of the random move has remained open to different interpretations. Some find it necessary to have at least two possible moves in order to describe a move as random, others prefer having a unique white move whilst some composers are ready to accept even an imaginary move as random.

This series of TC articles in two languages (simultaneously published in Russian in Problemist Ukraini and in English) began with a very ambitious goal - to bring closer different views about the concept of Threat Correction. Three composers, from different traditions and languages, tried to embrace a single TC theory. The process proved that our ultimate goal had been too ambitious. Instead, we have learnt more about our different opinions and decided to explain them publicly. We came closer to understanding each other with the hope that readers would come to their own conclusions. I will use this part of the series to express some of my personal feelings about TC and our joint adventure.

Back in 1977 there was a composing match, USSR - Yugoslavia. Barry Barnes, as the judge of the #2 section, suggested the pattern of thematic threats that now bears his name: 1.X? - 2.A/2.B, 1.Y? - 2.A and 1.Z! - 2.B. The USSR team convincingly won both the #2 and #3 sections, but first place in the #2 section went to our team, thanks to Milan Velimirovic and his No 30.

No 30. 1.Se~? – 2.Qe5, Qe6#. 1...Be4 2.Qc4#, 1...Re1 2.S:b6#, 1...Bf1!; 1.Sd3!? – 2.Qe5# (2.Qe6?), 1...Re1 2.Sb4# (1...Q:f4 2.S:f4#), 1...Be4!; 1.Sc4! – 2.Qe6# (2.Qe5?), 1...Be4 2.Se3#, 1...Re1 2.S4:b6# (1...Qh6 2.Q:g2#).

As Barry pointed out in his award, this 3rd degree white correction approached the theme in the way he had expected. The elegant mechanism of simultaneous closing of lines, especially the fl-c4-a6 line, offers a choice of threats and a wealth of changed mates, so very much in the style beloved by its composer.

The other top places were also dominated by a WC approach, using moves of a single white piece to unite the content. In No 31 we see the separation of three primary threats. At the end, both missing thematic threats reappear as variations, and you may recognize the well known Hannelius pattern used in many examples of Tertiary Threat Correction. The logic of TC is obvious and very close to that of TTC, albeit without the gradual introduction of the thematic threats that is essential for TTC.

No 31. 1.Se~? – 2.Qe5, Qd4, Qg4#, 1...Qa1!; 1.Sg4!? – 2.Qe5# (2.Qd4?, Qg4?), 1...Qa1 2.S:f2#, 1...Bg3!; 1.S:d3!? – 2.Qd4# (2.Qe5?, Qg4?), 1...Qa1 2.S:f2#, 1...B:e3!; 1.S:f3! – 2.Qg4# (2.Qd4?, Qe5?), 1...B:e3 2.Qe5# (2.Qd4?), 1...Bg3 2.Qd4# (2.Qe5?), 1...Qa1 2.S:d2#.

These two examples should help to explain my surprise at the treatment of the Barnes pattern that emerged after 1977. One might expect composers to continue to popularize the sophisticated methods of correction play used by the winners of the match. Instead we have seen hundreds of later examples with at least one artificial try by a second white piece, added only to complete the demands of the Barnes framework. One of the reasons for such a downgrading of the Barnes concept was its translation into the faceless letters of the "alphabetic formula".

It is often easier to find the key rather than the thematic tries where they don't belong to the same mechanism. This is why I became fond of the TC form - when different threats are needed - and WC form - when the same threat is presented in all phases. While being more difficult to use the same key-piece for different threats, the TC form allows stronger unity and easier understanding. No 32 was an attempt to present the so-called Ideal Dombrovskis in the form of doubled TC.

No 32. 1.Sc~(Sa2)? – 2.Re1#, 1...B:d4 2.Qf5#, 1...f3!; 1.Sb3!? – 2.Q:c6# (2.Re1?), 1...f3 2.Qf5#, 1...B:d4!; 1.Sd3! – 2.Sf2# (2.Re1?), 1...f3 2.Re1#, 1...B:d4 2.Q:c6#.

Some sources added another "thematic" try: 1.Se2? - 2.Qxc6#/2.Qf5#, 1...Rxc5! In reality 1.Se2? was a potentially nasty cook and not a part of the author's intention!

Only very recently I saw another composer combining TC with the Dombrovskis theme, but we will return to that later...

Coming back to the development of the Barnes pattern, composers soon managed to incorporate the ambitious le Grand theme. Once again, it took some considerable time before it was dressed in the demanding cloth of correction. Philippe Robert was among the pioneers; No 33 is one of several renderings he published in 1996.

No 33. 1.Sd~? – 2.Q:a1, Qd3#, 1...S:e4!; 1.Sb4!? – 2.Qd3# (2.Q:a1?), 1...B:c4 2.Q:a1#, 1...S:e4 2.Sc2#, 1...Sf4!; 1.S:e5! – 2.Q:a1# (2.Qd3?), 1...B:c4 2.Qd3#, 1...S:e4 2.Sf3# (1...K:e5 2.Qd6#, 1...R:c4 2.Qc5#).

This pleasant mixture of WC and TC logic, with complete Sushkov threat avoidance, may be compared with No 31. White corrects his random move to provide for 1...Se4!, and at the same time he twice corrects (reduces) the primary double threat, bringing back the "missing mate" in each phase. Please note how the well­organized play by a single white piece serves to assist the general construction of the phases, whatever the "main" theme may be. In my opinion, there is no main theme here. The ingredients include WC, TC, Barnes, Sushkov and Le Grand themes; these overlap and produce a strong impression. However, if one of the phases began with a piece other than the mighty knight, the magic would be ruined.

Through collaboration with Valery Shanshin I have learnt that it is not obligatory for the first moves to be made by the same white piece in the understanding of TC in Soviet literature. To describe the western meaning of TC, one has to use two terms: threat correction + choice of the key-move. This seems to be a case of misconceived translation, from the times when early western investigations in TC were transferred to the East.

There is a further potential of TC that I find inspiring - its relation to the theory of the Logical school (in #3 and #n). The reappearance of the primary threat may be perceived as the main-plan (Hauptplan), while the corrected threat has the role of the fore-plan (Vorplan). In order to execute the main-plan, it must be delayed in order for Black to complete the necessary conditions. Does not each Dombrovskis or Hannelius contain such characteristics too? Yes, but the logic appears deeper when you upgrade your plans while holding in your hand the same white piece! The overall impression gets even stronger when achieved with Black Correction, as in No 34.

No 34. 1.Sf~? – 2.Qh4#, 1...Be5 2.R:e5#, 1...g2!; 1.Sg2!? – 2.Sd2# (2.Qh4?), 1...R:e3 2.Qh4#, 1...Bc3!; 1.Sd5! – 2.S:f6# (2.Qh4?, Sd2?), 1...Bf~ 2.Qh4#, 1...Be5 2.Sd2#.

This rare combination of complete TTC and BC was the subject of very different evaluations by 5 judging countries in the WCCT. Our team gave it 3.5 points, but lower marks by the other countries indicated that the concept of TTC wasn't equally popular there. To remind the reader, Christopher Reeves formulated TTC in 2004 and it was quickly popularized by other British composers, but the language barriers seemed to restrict the new fashion.

In the following WCCT the marks given to No 35 brought me to a similar conclusion. Our team gave it the maximum of 4 points, praising the unity achieved by the same piece introducing all 5 phases and the same square for both thematic mates. Two central phases (unmasking black lines to prevent wQ's threats) end with reappearance of the avoided threats in the manner of TC and there is a wealth of transferred mates. Once again we differed in some general opinions, especially concerning the unity of the mechanism.

No 35. 1.Qd2? – 2.Re5#, 1...R:f4 2.Q:f4#, 1...e5(:f6) 2.Sd6#, 1...Qa5(e1) 2.S:d4#, 1...Re3!; 1.Q:d4!? – 2.Re5# (2.Qe5?), 1...Re3 2.S:e3#, 1...Qe1 2.Qe5#, 1...Q:d4 2.S:d4#, 1...Qa5!; 1.Qb8? – 2.Qe5# (2.Re5?), 1...Sc7!; 1.Qd6? – 2.Qe5# (2.Re5?), 1...e:f6 2.Q:f6#, 1...e:d6 2.S:d6#, 1...e5!; 1.Q:e7! – 2.Sd6# (2.Qe5?), 1...R:f4 2.Re5#, 1...Rd8(:c8) 2.Qe5# (1...R:e7+ 2.S:e7#).

During the last decade, I began paying more attention to judges' comments in the field of correction play and noticed that they often fail to mention some complex combinations. Take the case of three-mover No 36, an incredibly light presentation of Quaternary Correction by a single black piece (impossible in #2).

No 36. 1.Ra5! – 2.R:a4+ Sb4 3.R:b4#, 1...B~ 2.Qc2+ Kb4 3.R:a4# и 2...K:d4 3.Sf5#, 1...Bc6!? 2.Sc8 ~ 3.Sb6# и 2...Rb1 3.Be2#, 1...Be6!!? 2.S:e8 ~ 3.Sd6# и 2...Sc5 3.R:c5#, 1...Be4!!!? 2.Be6+ Bd5 3.B:d5#.

Let us see how 4 black errors accumulate, one by one:
- random vacation of d5 opens a5-e5 line;
- 1...Bc6 corrects it and adds anticipatory closing of e1-e6-b6 broken line;
- 1...Be6 repeats and corrects both while preventing black from using the e1-e6-d6 line;
- finally 1...Be4 repeats all three errors, corrects them, and adds closure of e1-e6 line.

The judge obviously missed the logic of the very difficult pattern. He mentioned only "black correction by BB" and criticized the last variation for being different. All these cases of different approaches toward correction themes illustrate how they enjoy very different popularity in different parts of the world. This was the reason to use two languages and two magazines for our articles.

Perhaps it would be useful to stress one general difference between White Correction and Black Correction on the one hand as compared to Threat Correction on the other hand. In the 3rd degree WC of No 30, White accumulates errors of the random move and the closure of e2-c4 diagonal. In the 4th degree BC of No 36, Black accumulates the errors of closing his lines. However, the 3rd degree TC of No 34 accumulates the positive effects of opening the h4-e4 line and guarding e3 which makes it substantially different.

Is correction an expression of form or content? I would say it may be a form but it may be the content, especially in its complex combinations such as TTC. There are many other ways to give TC thematic importance. In No 37 all three TC phases lead to the same flight-giving and changed mates.

No 37. 1.B6~(Bh1)? – 2.Kc6#, 1...Se7!; 1.Be8!? – 2.Bh5# (2.Kc6?), 1...f:g1~ 2.Kc6# (1...Kf3 2.Bh5#), 1...Sh6!; 1.Sb~? – 2.Kb4#, 1...Sc2!; 1.Sd5!? – 2.Sf4# (2.Kb4?), 1...f:g1~ 2.Kb4# (1...Kf3 2.Sf4#), 1...Bc7!; 1.Sc~? – 2.Kc5#, 1...Sb3!; 1.Se4! – 2.R:f2# (2.Kc5?), 1...f:g1~ 2.Kc5# (1...Kf3 2.R:f2#).

Thanks to David Shire and Valery Shanshin, this series of articles turned into much more than I expected at the end of 2013. It was Wieland Bruch who helped us to get an insight into the origins of TC from the late 1940s and a better picture gradually emerged. Many times I shared David's pleasure at unearthing valuable examples from the past. Led by instinct and a deep appreciation for our art, David has been proving that most of the modern concepts of TC had been intuitively presented long ago without any formulated theory. It seems TC ideas have never been highly fashionable, but they kept intriguing the creative minds of an earlier era, especially in the West.

One of David's recent surprising discoveries, No 38, partially anticipates our own joint #2 with Christopher Reeves, whose version was included in Part 2 of these articles as No 20. With 1.e4? substituting for a random move by wSd5, the Danish master presented two TC phases. He used a very surprising solution for the fourth transference of the thematic 2.d3 and in the process added a Dombrovskjs paradox. David found many more TC rarities and I hope that he will have an opportunity to present some of these.

No 38. 1...B:d5 2.Sd3#, 1...Q(S):d6 2.Qf6#; 1.e4? – 2.Sd3#, 1...b2 2.Bd4#, 1...f:e4 2.Qf4#, 1...Sc5!; 1.Se3!? – 2.Q:f5# (2.Sd3?), 1...Bd4 2.Sd3#, 1...S:d6!; 1.Se7!? – 2.Q:f5# (2.Sd3?), 1...S:d6 2.Sd3#, 1...Bd4!; 1.e7! – 2.Qf6# (2.Sd3?), 1...Q:d6 2.Sd3#, 1...f4+ 2.Q:f4# (1...K:d6 2.e8Q#).

Valery was the initiator and the motor of our work. In what seems to be the most fertile period of his composing, he enriched the field of TC with many new ideas during our collaboration. It is a real pleasure to end my short part of our story with his latest TTC-Dombrovskis combination, No 39, announced in the comment to No 32. With some artistic freedom in presenting the random move, the whole composition became complex and lively at the same time, incorporating cyclic pseudo-le Grand as an additional link between the phases.

No 39. 1.Bh8? – 2.Rd4#, 1...f4 2.Qd3#, 1...e5!; 1.Be5!? – 2.Qd3# (2.Rd4?), 1...B:b4 2.Qf4#, 1...Qf3!; 1.Be3! – 2.Qf4# (2.Rd4?, Qd3?), 1...e5 2.Rd4#, 1...Qf3 2.Qd3# (1...B:b7 2.Q:f5#, 1...B:e3 2.Sc3#, 1...Se2(h3) 2.Q:h1#).

Be sure; Valery has a lot to say and present in Part 4, planned for November 2015!