Harald Keilhack Bare ukuce cus RCHarald Keilhack Knight on the Left: 1.Ac3 Fully updated edition Occupying the territory where unorthodox and main-stream openings meet, the Knight on the Left can undoubily be a forceful weapon: The Dane Ove Ekebjeerg nearly became correspondence World Champion, playing 1.23 exclusively! This is the first full-length book about this interesting opening, unfolding a complete theory of 1.44c3; including several off-beat lines against semi-open defences like the Alekhine or the Caro-Kann. The author thoroughly explains the basic structures which supply huge scope for creativity. There is broad discussion of move order subtleties and transposition issues, so that the book deepens your understanding of opening play in general! The author, FM Harald Keilhack, first attracted attention with his 1994 book on the Tarrasch Defence. He works as editor of the Schachverlag Kania and as chess correspondent for the Stutigarter Zeitung. For this English language edition he has thoroughly updated the original German manuscript, published in 2003. “If you have a liking for the unusual and daring, you will probably not find many books this year that are better than this one!” (Carsten Hansen on Chesscafe, Checkpoint September 2003) “A marvelous investigation of the eccentric (so far!) development of the queen's knight on the first move ... | highly recommend this book for anyone who wants to get off the beaten path.” (John Watson, The Week in Chess, Book Reviews #56) “| hope that both regular and alternative players will read and enjoy this book. | myself certainly did, and not only because it was a pleasure to see the concepts | had developed so long ago worked out in so many different ways. The most important thing is that this book may stimulate more players to play this opening, for | know that it can give a lot of play- ing pleasure and fun.” (Dick van Geet, New in Chess Yearbook #68) €24.95 £17.99 $30.95Harald Keilhack Knight on the Left: 1.2\c3 Studies of an Unorthodox Chess Opening K y 2 Ip Schachverlag Schachverlag KaniaISBN 3-931192-29-6 1th Print © 2005 by: Schachverlag Kania Originally published in German language as “Der Linksspringer 1.2\c3”, ISBN 3-931192-20-2 © 2003 by: Schachverlag Kania Address: Schachverlag Kania Richard-Wagner-Str. 43 D-71701 Schwieberdingen Germany Tel./Fax +49 (0) 7150/3 7098 eMail: Kaniaverlag@yahoo.de, Keilhack@aol.com Internet: www.kaniaverlag.de Typeset: Harald Keilhack Translation: Harald Keilhack Proofreading: John Adams Print & Binding: Druck Viener, Vimperk/Cz Cover design: Frank Stiefel All rights reservedContents 3 Contents Contents Symbols Preface Acknowledgements .. Chapter! 1.4\c3 e5. Chapter Il The Van Geet Attack 1.4\c3 d5 2.e4 d4 3.A\ce2. Chapter Ill 1.2\c3 d5 2.e4 dxe4 3.2\xe4. Chapter IV The False Semi-Open Games — 1.2c3 d5 2.e4 c6, 2...e6 and 2...Af6 .... Chapter V Against the Sicilian - 1.2\c3 c5 2.413... Chapter VI Alternative Systems after 1...e5/...d5/...c5. Chapter Vil Various Answers to 1.4)c3 ... Index of Players .... Index of Variations Sources4 Symbols Symbols +- White has a decisive advantage + White stands clearly better £ White stands slightly better = balanced position eo unclear position + Black stands slightly better ¥ Black stands clearly better -+ Black has a decisive advantage a with compensation for the material A with the idea (or threatening) x weak point # mate " brilliant move ! good move 12 interesting move 7 dubious move ? bad move 2? blunder (ol) olympiad (ch) championship corr. correspondence game Wch, Ech World or European Championship respectively (mostly correspondence) mem memorial (mostly correspondence) 2 see or leads toPreface 5 Preface This book covers the opening 1.2c3. Opening? - Oh, yes! Many consider 1.43 to be just a move, which might be playable, but “hardly has any indepen- dent value”, rather transposing to 1.e4 systems. In this context, Khalifman’s five volume opus 1.@f3 - Opening for White accord- ing to Kramnik springs to mind. In an odyssey of more than 1300 pages, the former FIDE world champion proves that the “knight on the right” 1.23 is a good move, but not an opening in itself! He doesn't at all cover the Reti (with which 1.28 is usually connected), but a multi- tude of openings including the King's In- dian, the Queen's Gambit (Slav, Orthodox, Accepted etc), the Dutch Defence, the English Opening and even some lines of the Sicilian (namely the Maroczy setup against the Accelerated Dragon); besides, some peculiarities like the Grinfeld without d2-d4. Quintessentially, 1.23 is presented as a very subtle transposition tool, by which - to simplify a bit - undesir- able lines of the aforementioned openings can be excluded. So, how are matters after 1.2\¢3? There's a bunch of established independent con- cepts, e.g. after 1.503 &5 2.213 Acé 3.04 or 1.403 d5 2.e4 d4 3.A\ce2. It's mostly possible, but not always advisable to stay clear of all transpositions. Typical que: tions that a 1.4\c3 practitioner will face are: * After 1.23 d5 2.e4 e6, should he go for a French by 3.d4 or try some independent setup? © After 1.23 c5, should he play 2.e4 im- mediately, possibly going for a Closed Sicilian (3.93) or for the Grand Prix Attack (3.f4)? Or transpose to a stan- dard Sicilian by 2.f3 Dc6 (2...d6, 2...6) 3.04 cxd4 4.\xd4 and 5.e4? Or strive for the very tricky indepen- dent lines, keeping the option of e2-e4, transposing to some standard lines, as an “emergency exit” in reserve? * After 1.3 g6, should he return to 2.e4, should he go for the fierce 2.h4 or should he stay flexible by 2.2\f3 and 2.93, not yet revealing his intentions? In fact, every leading 1.4\c3 player has his own answers to those questions. Trans- positions at a later stage are full of subtle- ties, which will be revealed in this work. If ‘one wants to switch from 1.e4 to 1.23, this is even an advantage: you may start by playing 1.e4 e6/c6/A\f6 2.!c3 (reach- ing some lines from Chapter IV), later on you play 1.463, firstly transposing quite often to 1.e4 systems, then less and less. This book covers every possibility after 1.4\¢3 which at least has some indepen- dent value. Additionally, some “secret” lines which belong to 1.e4 openings but in practice occur more often via 1.4\¢3, or which have some features of this opening. For example, such Anti-Caro- Kann lines as 1.e4 c6 2.4\c3 d5 3.f4 or 3.413!?. Typically such lines are neglected in opening manuals. A special focus is on late transpositions (e.g., on move 6) to 1.e4 positions under favourable circumstances! | have excluded only 1.23 d5 2.d4, which leads to the Veresov Opening after 2... 46 3.295, to the French after 2...e6 3.e4 or the Caro-Kann after 2...c6 3.e4. Besides transpositions from 1.4\c3 to 1.e4 openings, there are also some in the other direction, e.g. 1.e4 e5 2.43 d5 3.4\c3!?, 1.e4 d5 2.5\03!?, 1.44 BE 2.43 5 3.23 (a basic position from Chapter V); one can argue that 1.e4 c5 2/4 d5 3.4c3 is also truly a 1.A\c3 position.6 Preface An unrecognised opening Unlike unorthodox openings such as 1.b4 or 1.94, 1.2¢3 by no means com- promises White's position, and at the same time it doesn't limit the first player to some extravagant strategy or other. On the contrary, this move is in perfect harmony with classical principles of opening play like development and gaining influence on the centre (keeping an eye on e4 and d5). Therefore, it sounds strange that 1.23 is not yet recognised as an opening in its own right. Admittedly, the c-pawn is blocked, but in every opening, every move kills off some option or other. Actually, 1.23 offers exceptionally good possibilities of winning a miniature, thanks to knight tactics and supported by the accelerated development of the minor pieces. The older 1.\c3 literature almost reads like a collection of traps and miniatures. However, the general level of chess is increasing, and one cannot expect that the average player of our day will repeat the errors that some semi-celebrities of the past committed - not to mention modern computer-aided correspondence chess. So, this book isn't at all about cheap traps but will establish a deep theoretical and strategic basis for the further ex- ploration of the Knight on the Left. Between Worlds | was attracted by the unique hybrid stat- us of a so-called ‘uriorthodox” opening which has nevertheless so many lines which connect to traditional openings. 1.4c3 is still labelled as an “unorthodox” opening and accordingly, it’s treated poorly in main-stream publications like Encyclopedia or Chess Informant. On the other hand, many positions which arise can only be reliably assessed when one compares them carefully to related, but well-explored positions from standard openings. Often enough, en- thusiastic amateur authors of books about unorthodox or gambit openings are lacking in general knowledge of standard positions, which results in blatant misjudgements. In this book, | do my best to transfer knowledge between these two, so different, worlds. This also includes, by the way, comparisons to Positions with colours reversed, which often allow striking insights. Some methodological issues: | agree very much with the maxim that if a given position can occur in various places (via transposition), all relevant games and analysis have to be presented together. This necessarily leads to a huge number of cross references. When- ever various move orders can lead to the same position, | have tried to figure out which one is most precise for both parties - sometimes this wasn't easy at all (e.g. in Section 17). Unfortunately, for instance, Dunnington’s book is rather sloppy in this respect. “Between Worlds” was also a maxim when choosing games from corre- spondence as well from over-the-board chess, from grandmaster as well as from amateur practice. This was the case too in my methodological choice to keep the sample game approach (with deep strategic explanations), but at the same time to supply dense theoretical coverage. The fact that this book has expanded to nearly 400 pages should not distract from the fact that even with today’s knowledge some fundamental issues around move 5 are still unresolved. With regard to possible future discov- eries | haven't cut off lines that have to be regarded as unfavourable from today’s point of view.Preface 7 History and Naming 1.43 was first mentioned in the 1st edition (1843) of “Bilguer - Handbuch des Schachspiels” by Tassilo von Heydebrand und der Lasa. According to Adrian Harvey/Stefan Bicker (Kaissiber 15), the first recorded 1.4:c3 game was Kieseritzky-Desloges, France 1847 (p.21). The game Napoleon-Madame de Remusat, Malmaison Castle 1804 (p. 10), is probably a hoax. Later on, 1.403 was occasionally tried by world class players, e.g. Blackburne- Noa, London 1883, or Bogoljubow- Zubarev, St.Petersburg 1925. But these were solitary cases, comparable to Morozevich-Kasparov, Frankfurt 2000 (see Game 83) in our own days. It is not the professionals, but rather enthusiastic amateurs and grandmasters of corre- spondence chess who have advanced the development of this opening. Thus, about 1910 the Czech Johann Kotre started to play 1.c3 regularly, as did some years later Ted Dunst from New York. After 1945, the Baden master Leonhard Hanke attracted attention by playing 1.\c3 in southern Germany. Further progress is closely connected with the contemporary heroes mentioned below. The opening 1.2c3 has had many names; it is possible to say that every- body who wrote about it has used his own name. So, there had been, among others, the Romanian, Kotrc, Dunst, Sleipner and the van Geet Openings (to mention just the most widespread ones). Finally, | opted for the neutral Linksspringer or Knight on the Left, which was preferred by my mentor in chess publishing, the Randspringer editor Rainer Schlenker. Contemporary heroes From about 1965 up to nowadays, two outstanding correspondence GMs have been setting the trend: Ove Ekebjerg, Denmark, and Dick van Geet from the Netherlands. The extrovert Dutchman has published articles, booklets and a CD about the “Van Geet Opening’. In his work, tactical finesses prevail. Of the reclusive Dane, however, hardly anything is known beyond the game scores them- selves. But his games are of unique strategic depth. In the 14th correspondence world cham- pionship 1994-2000, Ekebjzerg' came close to his greatest triumph: he scored 10% out of 14, including 6 out of 7 in his games with White, opened exclusively with 1.2c3, and almost looked like the sure winner. However, in the end Tinu Yim from Estonia, who finished his games years later due to longer postal delivery times, overtook him by half a point. Besides van Geet and Ekebjzrg, other correspondence GMs including Sarink, van Perlo and Hector regularly start their games with 1.43 (and there are only about 200 corr. GMs); and besides them, numerous corr. players around 2400 too - one could almost talk about a “corre- spondence chess opening’! In over-the-board chess, one will repeat- edly meet names like, IM Mestrovic, IM Sydor and (again!) IM van Geet, see for instance his draw against the then world champion Spassky (Game 14). Nowadays, 1.2c3 is occasionally used by modern grandmasters like Bellon, Buhmann (soon to be a GM), Ermenkov, Hector ‘In the years since publishing the German edition, Ekebjzerg’s fortunes have turned. | have in mind his failures in the 16th World Championship and the 50th Jubilee Grand- master Tournament (a true all-star correspon- dence event). However, those results have been influenced by health problems.8 Preface (again! - he has both titles), Cs.Horvath, Narciso, Rashkovsky, Rogers, Schmaltz (limited to 1.e4 d5 2.2\¢3), Wi.Watson; all of these have a well-earned reputation as unprejudiced players. Other names include Dutch followers of van Geet, namely Jongsma, Leeners, van Bellen; the 1.2c3 scene from Tubingen (Frick, Schlenker, Moser and others) or the Estonian master Aarne Hermlin. And finally Anker Aasum from Norway. His book, while being somewhat dated with regard to pure theoretical issues, contains a lot more historical information, including numerous short biographies of enthusiastic 1.3 players (or rather 1.4)c3 riders, as Aasum uses to say). Acknowledgements | would like to thank: Jochen Bastian, Christian Beyer, Jonathan Tait and Martin Weise for sending me their private game collections. Gerson Berlinger, Petrus Burghouwt, Gabriel Elefteriadis (who painstakingly pointed out some inconsistencies in the German edition), Wolfgang Finke, Tihomir Glowatzky, Bernd Grafrath, David Héffer, Marc Lacrosse, Tobias Lagemann, Ralph Mallée, Davide Rozzoni, Rainer Stock- mann, Dick van Geet, Martin Weise, Frank Zeller, who submitted games, analysis or general comments which have been useful for this revised edition. Henk van Bellen for the kind invitation to his nicely organised thematic tournament in Mariac 2001 (also including van Geet, Aasum, Hanke). Stefan Bicker for historical references. And finally John (lan) Adams for the proofreading, ensuring that this experi- ment with an “author's translation” didn't turn out too badly, or so | hope. Harald Keilhack Schwieberdingen, March 2003/August 2005 About this revised English language edition Soon after the German edition of this book was published in Spring 2003, | received several requests for an English edition. Well, this project took some time, so that simultaneously the question of an update arose. In brief, this edition is totally revised in the respect that | have included all im- portant games played until August 2005 (e.g. Morozevich-Kohler). On the other hand, | have kept the whole structure of the book, including the 99 sample games, untouched. Partly for practical reasons, having the countiess cross-references in mind. But besides this, | felt that those 99 stem games perfectly illustrated the original concept of the specific systems and variations - for modern refinements, see the more recent games and analysis in the notes. Also, some errors and misjudgements have been corrected. There are many more new games than it may look like at first sight, as on correspondence games mostly only the starting date is published, so “corr. 2001" reads: started in 2001, finished in 2004 (or so) and published only in 2005, when the tournament concluded. However, | haven't included material from the recent Chessbase CD on 1.4\c3 - Carsten Hansen's Chesscafe review should give you an idea why. (www.chesscafe.com/text/hansen73.pdf) Finally, although much recent stuff has been added, this edition is actually 16 Pages smaller than the original one. Apart from the fact that the German language takes up more space, | took van Geet's advice in his review (New in Chess Yearbook 68) to leave out some non- essential things. Harald Keilhack Schwieberdingen, August 20051.23 e5 9 oO hapter | — 1.4)c3 e5 The answer 1...e5 in the spirit of the open games is pretty much the best the 1.4c3 player can hope for. With 2.2f3!, White preserves at the very least the advantage of the first move, and in many cases he can hope for much more. For more specific introductions, | refer you to the separate sections and games. First, an overview: Section 1: 1.c3 €5 2.043 d6 3.d4 Section 2: 1.23 e5 2.53 205 (2...d5) Section 3: 1.2\¢3 e5 2.263 cb 3.04 exd4 4.2\xd4 various Section 4: 1.23 e5 2.43 Dcb 3.04 exd4 4.Axd4 Dxd4 5.Wxd4 Section 5: 1.23 e5 2.213 Acé 3.d4 exd4 4.axd4 AE 5.295 Section 6: 1.23 e5 2.23 “\c6 3.d4 exd4 4.2\xd4 2b4 Section 7: 1.203 e5 2.243 Dcb 3.d4 exd4 4.axd4 &c5 i but before that we will look at some Section 1 gambit lines in Section 2. 1.2\c3 e5 2.013 d6 3.d4 oe If one assumes that it will become a Philidor Defence in any case, then 3.e4 Game 1 would come into consideration too - Schlenker — J. Pape Villingen (rapid) 1993 1.203 e5 2.013 The best continuation: development along with an attack on e5. We'll examine the alternative 2.d4!? in Section 35. 2...d6 In the spirit of the Philidor Defence. The most natural and most frequently played continuation is 2...¢6 (Section 3-7), Black will probably continue with 3...2'6 4.d4 2bd7. White's hesitation - see the moves 1.e4 e5 2.213 d6, and now (instead of 3.d4) 3.24c3 first -, could motivate Black to try deviations like 3...f5!? or 3...2.g4!. This leads us to the oldest 1.2:c3 game 1.23 e5 2.213 d6 3.e4 5 4.n3 (4.04 or 4.204 fxe4?! 5.Axe5!, Moser-Fuchs, Tubingen 1987) 4....xe4 5.Axe4 Acé6 6. Ng5 d5 7.WhS+ g6 8.813 Dh 9.A\e+ Be7 10.2xd5+ Sd6 11.e4+ wxdd 12.2.04+ Sxc4 13.Wb3+ Ld4 14.Wd3#,10 Chapter | Napoleon-Madame de Remusat, Malmaison Castle 1804. However, the authenticity of this game is doubtful. But 3...2g4! seems to promise easy equality: 4.2.64 (or 4.2e2) 4...2.¢6, and suddenly it's not so easy for White to accomplish d2-d4. In Bjarnason-Dittler, Bad Wérishofen 2001, Black deviated with the seemingly safe 4...&h5? (pro- tecting 17, instead of 4...2c6) and was punished by 5.2\xe5! £&xd1? (5...Wg5! 6.d4! Wxe5 7.94 We7 8.gxh5+) 6.2xf7+ Be7 7.2054. Relatively best seems to me 4.ad5!? (A&e3) from the game Minz-Fihrer, Rheinhessen 2000. All these are - and by far not for the last time in this book - subtleties in a grey area of the opening theory. Nobody has considered it worth the trouble to examine whether Black may exploit the ‘inaccuracy’ 1.64 e5 2.43 d6 3.43 or not. The Encyclopedia only mentions some not really convincing transpositional possibilities. After 3.d4, Black has three moves at his disposal: (see diagram on the next column) 3...exd4 Besides this, 3...2c6: there are 3...2\d7 and 93!?, and 4...Agf6 5.292 e4!? is an oddity, van Geet-Burghouwt, Cannes thematic 2004, 6.295 d5 7.f3 h6, and now 8.h3 exf3 is more or less equal. @l lack may vary with 4...c6 or 4...2e7, but this doesn't make much difference), reaching the basic position of the Philidor Defence (1.e4 e5 2.3 d6 3.d4 Df6 4.203 Dbd7). May we consider this as a success for White? - Well, it's not possible to ‘refute’ the rock-solid Philidor Defence, but its reputation is to be somewhat passive, which causes some theoreticians to apply the verdict ‘+’ to it. Instead of the routine 5.2.04 2e7 6.0-0 0-0 7.2et (7.We2, 7.24) 7...c6 etc, where the Philidor devotee will feel at home, there are three suggestions for a setup with g2- 93 or g2-94 (which doesn't make much of a difference, as will soon become clear): a1) 5.h3!? £e7 6.94 c6 7.2g2 h5 8.95 Dh? 9.h4 DAhfB 10.2e3 Ags 11.Wd2 DdB 12.De2 294 13.Dfg1 16 14.43 e6 15.d5 cxd5 16.exd5 &d7 17.0-0-0 @Dh7 18.4 exf4 19.4\xf4 Dxf4 20..2xf4 fxg5 21.hxg5 0-0 22..Ne2 294 23.2h3 and finally -¥2/37 in the game Klinger- Wahls, Bad Wérishofen 1988, but White had his chances throughout the game. a2) Recently, even the uncompromising 5.941?1.23 e5 11 has become popular, e.g. 5...2xg4 5...g6 6.95 (or 6.2c4) 6...Ah5 7.263 2g7 8.Wd2 0-0 9.0-0-0 5 10.exfS Sxt5 11.dxe5 Axe5 12.0xe5 2xe5 13.2044 with the initiative, Lastin-Azmaiparashvili, Tripoli 2004. 6.291 Agté Perhaps better is 6...exd4 7.xd4 Ade5. 7.2.04 h6 8.2.3 c6 9.Wd3 9.dxe5 dxe5 10.Wd3 Dh5 11.2xf7+ Sxt7 12.0xe5+ Dxe5! 13.Wxd8 AVS+ 14.801! Axgt 15.%c1/t, Shirov- Shaw, Gibraltar 2005. 9...We7 10.0-0-0% b5?! 11.2xb5! cxbS 12.dxe5 dxe5 13.2\xb5 Wa5 14.Wed with a promising sacrificial attack, Shirov-Azmaiparashvili, Plovdiv 2003. For more details on this line see the New in Chess Yearbook #75. In fact, this is a slightly improved version of 1.53 Af6 2.94 (see Game 90) - which is generally considered as a typical amateurish gambit, having nothing to do with so-called “serious” chess. Sometimes the gap between the two becomes very narrow! a3) 5.g3!? 2e7 6.292 0-0 7.0-0 c6 8.a4 a5 9.h3 He8 10.2e3 2f8 11.Be1 We7 12.Ad2 b6 13.f4 &£a6 14.94 with good prospects for White, Lach-Bachler, Germany 1997. Instead of 5....£e7, Black may also con- tinue with 5...g6 6.292 297 7.0-0 0-0. A brief introduction to the “g3-Pir This brings us to a digression This quiet, primarily positionally orientated system against the Pirc Defence starts as follows: 1.64 d6 2.d4 Af6 3.2¢3 g6 4.93!? 297 5.292 0-0 6.ge2! But not 4g1-f3. A possible ....294 may be annoying, but besides this, a formation with “\e2 is much more harmonious: f2- {4 is possible as well as 3-94, De2-g3. Anyway, proceeding with 93-94 is rather typical for that system - as we have al- ready seen in an accelerated form in the Klinger-Wahls game. White wants to maintain the d4-square (and he is able to do this!), an early d4- d5 or d4xe5 is not desired. Therefore, in the case of 1.e4 g6 2.d4 297 3.23 dé, the move 4.@ge2! recommends itself, because after 4.93 Black can prevent White's setup by playing 4...2\c6! 5.23 (5.2ge2 294) 5...e5: 6.dge2?! g4!. 6...e5 7.h3! Continuing prophylaxis with regard to d4! Because after 7.0-0 “\c6!, White has some troubles at this spot: 8.2e3? Qg4 or 8.h3?! exd4 9.Axd4 2xe4!, and the following series of exchanges gives Black equality at least, while 8.d5 or 8. dxe5 is not exactly in White’s intentions After 7.3, the basic position of the system is reached:12 Chapter | If Black continues in the usual slow manner of the Pirc Defence (e.g. 7...c6 8.a4 Dbd7 9.2e3 He8 10.0-0), White preserves the more active possibilities on both wings, whereas in the centre the usually so critical points d4 and e4 are safely protected. The best continuation for Black is the straightforward 7...2\c6 8.2.3 Be8 9.0-0 exd4 10.xd4 2d7. Back to the last but one diagram po- sition. Obviously, both parties must make concessions compared to their optimal setup. White would like to have his knight on e2, but on the other hand he has saved the time for h2-h3. Thanks to ...bd7, neither ...2g4 nor the other outlined possibilities for counter-play have to be feared. If Black finally plays ..exd4, the detail which works to White’s disadvantage disappears after f3xd4. Therefore in the game Andersson- Shirov, Ter Apel 1997, Black developed his queenside laboriously but still solidly with 8.h3 c6 9. a4 We7 10.23 b6 11 Wd2 &b7 etc. I do not want to give a final verdict about the last two diagram positions, but to sharpen the reader's eye for such trans- Positional possibilities and their assess- ment. Paradoxically, the ‘perfect’ 1.2\c3 player needs an especially good general opening knowledge, due to the many transpositional possibilities. Often a transposition is by far the best option, particularly if you receive a small bonus (e.g., if Black has already forfeited the chance for the best setup in the par- ticular system, or some critical lines are already avoided). This makes the big difference that separates 1.3 from, say, 1.64. But don’t be afraid - in the most transpositional cases, it is not ne- cessary to know any complicated forced lines. Rather it is basic knowledge which is required, e.g. which setups are the best ones for both parties in particular systems. Saying this, we finish this digression. We'll use the last diagram position as a point for future reference; and maybe some 1.4\c3 players or other will become familiar with this system which is not yet popular in 1.A\c3 circles. Especially with regard to various cases in Chapter Vil. b) 3...Ae6, and in accordance with his taste, White can choose one of three roughly equal continuations: b1) 4.e4, a blend of the Philidor Defence and the Three Knights Game with pos- sible transitions to the Ruy Lopez. The Encyclopedia assesses (under C 46) the sequence 4...2g4 (4..exd4 5.2\xd4 3ad7 - 5...f6 6.2b5 may lead to a Ruy Lopez - 6.2e3 Sf6 7.2e2 g6 8.Wd2 2g7 9.0-0-0 0-0 10.13 a6 11.94 with an edge, Radulov-Westerinen, Hamburg 1981) 5.dxe5 (5.d5 id4! is unclear, 6.231? Axi3+ 7.gxf3 &d7 8.Wd2 with mutual chances) 5...dxe5 6.\W/xd8+ Exd8 7.205 2b4 8.2xc6+ bxc6 9.263 2xf3 10.gxf3 £5 11.exf5 as ‘2’, and furthermore it refers to a transposition to the Ruy Lopez after 5.2.65 (1.e4 e5 2.213 Acé 3.205 d6 4.d4 294 5.2034, C 62), The databases are rather overcrowded with games in this practically unexplored no man's land of opening theory. b2) 4.d5 Ace7 Now Black is playing a ‘Left Knight’ with colours reversed. But in fact, he has1.23 e5 13 already limited his possibilities by play- ing ...d6: no ...\g6/...25. The position after 5.e4 is usually reached after 1.e4 5 2.213 Acé (d6) 3.d4 dé (Ac) 4.05 @Qee7 5.403 or 1.64 cb 2.d4 e5 3.d5 Dee7 4.23 d6 5.23. | covered such lines in my book 1...2\¢6 aus allen La- gen (Keilhack/Schlenker, Kania 1995), but there you'll mainly find games in which White played c2-c4. This doesn't say anything about the value of c2-c4 (compare the discussion in Chapter II for this question). Two facts in brief: By the move ...d6, the second player is committed to a setup with ...g6/...297. © The move Ag1-13 also is not the best in this kind of position. We shall end this paragraph by pointing out that Black has tried very different continuations after 5.e4: namely 5...g6, 5...f6, 5...06, 5...294 and 5...f5. And that White may consider 5.g3 (instead of 5.64) as well. b3) 4.dxe5 is characterised by some unsolved details. After 4...g4!? we reach the vicinity of the Englund Gambit (1.d4 e5 2.dxe5 d6 3.293 2g4 or 2...Ac6 3.243 d6 4.exd6 &xd6). Whereas 4...dxe5 5.Wxd8+ @xd8 or 4...xe5 5.2\xe5 (maybe 5.e4 instead) 5..dxe5 6.Wxd8+ xd8 represents some obscure endgame line similar to 1.d4 d6 2.c4 e5 3.dxe5 dxe5 4.Wxd8+ Gxd8 (this is, thanks to the move c2-c4, which only weakens the d4-square and restricts the f1-bishop, considered rather advantageous for Black!) or 1.e4 d6 2.d4 e5 3.dxe5 dxe5 4.Wxd8+ w&xd8. | think that at least in the version with 4...2xe5, Black has rather safe equality, in view of the typical manoeuvre ...c6/ ...&c7; White also lacks the potential for an attack on the e5-square (often {2-4 followed by g1-f3 is strong in similar positions). 4.2xd4 4.206?! As was to be expected after 2...d6, we're still on the ‘Philidor track’. Therefore, a) 4.2) a1) 5.e4 leads to another basic position of this opening: © 1.e4 e5 2.013 d6 3.d4 ARG 4.2\c3 exd4 5.2\xd4 (5...2e7 or 5...96). White has two alternatives to 5.e4: a2) 5.295 2e7 6.2xt6!? 6.Wd2, 6.e3 (given a ‘? by corr. GM Mallée) 6...c6 7.Wd2 0-0 8.2d3 He8 9.0-0-0 b5 10.25 &xf5 11.2xf5 d5, Mallée-Wiemer, corr. 2002/03 with better attacking chances for Black. 6...£xf6 with knight tactics in W.Schmidt- Rothe, Borkum 1996: 7.Ad5 295 8. Db5 c6!? (8...a6 9.h4!) 9.\de7+ we7 10.\Wxd6+ Wxd6 11.axd6 Gxd6 12. ®xa8 2d8 13.93 2e6 14.292 Aaé, Black finally won after he picked up the a8-knight. One has the impression that White missed some better possibilities (9.Abce7+ as well as 10.2\xa8 are ob- vious alternatives); probably, this is a very intricate story. One general remark: in the very similar positions from Chapter V (Ae7 instead of &c7), Black's central superiority is a long-term strategic trump, so White should 0, if possible, for some early tactics. Here White can build up his game slowly,14 Chapter | limiting Black's counter-play and relying on the strength of his e- and f-pawns. a3) 5.93(!) 2e7 6.292 0-0 6...d5!?, in spite of the loss of a tempo. 7.04 (exactly this move order was played in Vingerling-A.Denk, Germany 1998), and due to the e7-bishop, Black has an inferior version of a g3-Pirc; his position offers no real prospects for counter- play. Occasionally the position under discussion has been reached via other roads, e.g. in the game Balashov-Lau, Dortmund 1993. At this point there fol- lowed 7...2e8 8.0-0 c6 9.h3 2f8 10.14 Dab 11.443 2e6 12.22, It's worth mentioning that in the Philidor Defence White cannot reach this position: 1.64 e5 2.213 d6 3.d4 @f6 4.03 exd4 5.Dxd4 2e7 6.93, and 6...d5! (6...294!?) equalizes on the spot (OK, recently some messy complications after 7.e5 g4 8.2f4 g5 9.e6 had been discussed) - the disadvantage of the ‘premature’ e2- e4! Instead of 6.93, 6.2\d4-e2 is prefer- red, for example 6...0-0 7.93 He8 8.292 28 9.0-0 a6 10.He1 Ac5, and now 11. e2-d4 in the game Bruzon-Nisipeanu, Decameron 2003 - White was happy to reach a position two tempi down com- pared to our line! Instead of 5...:2e7, in Schlenker-Kubicka, Linz 1988 there occurred 5...c5!? 6.b3 d5. There followed 7.292 (7.295!?) 7.806 (7...04!?) 8.295 Le7 (8...d4 9.2xb7) 9.2xf6 2xf6 10.2xd5 2xb2 11.0b1 23! (11...axd5 12.8xd5 203+ 13.611 Dd7 14.8xb7 Bb8 15.%c6+- Schlenker) 12.¢4 0-0 13.0-0 @\c6 14.4)a1! Eb8 15.2c2 &b4, and according to Schlenker, White should now play 16. exb4 exb4 17.2xb4 Wxd1 (17...axe4 18.Axc6 bxc6 19.We2 2d5 20.e4) 18. Efxd1 @xb4 19.Exb4 a5 20.Exb7 Sxb7 21.2xb7 &xc4 22.Hd2+. Nevertheless, the kind of position after 6...d5 is not so easy to classify in its proper place (with the c5-pawn on eS or e6 it would be a Griinfeld reversed), and therefore some questions remain unsolved. b) 4...d5 (tries to underline his influence in the centre at the cost of a tempo) 5.414 D6 6.e3 &b4 is roughly equal, Haagen Hansen-H.H.Schmidt, Copen- hagen 1998. 5.e4 (5...dxe4 6.s2b5) may lead to a small advantage in development. c) 4...g6, and 5.e4 £97 leads to another frequently played Philidor position: 1.64 5 2.218 d6 3.d4 exd4 4.2xd4 (probably a bit stronger is 4.Wxd4) 4...g6 5.23 297. Now White has the choice of several standard moves (6.2.¢3, 6.2.c4 6.22, 6.g3 etc). However, the only ex- ample cited in the Encyclopedia shows the avant-garde move 6.h4!? (6.016 7.295 h6 8.814 DAcé6 9. We2 a5 11.a4 0-0 12.0-0- @Qb4 14.Wd2 h5 15.2954, Istratescu- Barbero, Martigny 1993). Yes! - 1.¢3 players have always known it: ...g6 should be attacked with h2-h4, if necessary as early as on move two (1.3 g6 2.h4). Therefore, 5.h4 AIG 6.295 h6? (caused by exaggerated fear of 6...297 7.Ad5) 7.2xi6 Wxt6 8.205 Wd8 9.b5 Dab 10.Wd4 16 11.Ax16+ with a massacre, A.Volkmer-G.Neuber- ger, Germany 1996. Beside, White may head for a g3-Pirc by 5.93; and Black may put his own stamp on matters with 5...297 6.292 2e7!?, Pel-Ernst, Dieren 1998. d) 4...2e7 5.e4 (or 5.93!, compare under a3 above) 5...2f6 6.2¢3 £d7 7 Wd2 Ac6 8.b3 Age7 9.0-0-0 0-0 10. 2e2, and White has the easier game, Jendrossek-Mineo, corr. 1983. 5.93 The move 4...c6 is already somewhat unusual with regard to the Philidor Defence. In opening theory, the position1.2c3 e5 15 after 5.e4 is categorised not as a Phili- dor but as a Scotch Game with 1.e4 e5 2.43 Dc6 3.d4 exd4 4.axd4 dé (a passive deviation from the main lines with 4...205 or 4...2f6) 5.23. Also possible is §.\xc6. Black already has problems getting a playable position at all. The transition to a g3-Pirc turns out to be problematic due to unobstructed g2-bishop: ows 7...g6, and now 8.e4 2g7 9.h3 0-0 10.2.3 He8 would lead to a main line of the g3-Pirc: 1.64 d6 2.d4 f6 3.203 g6 4.93 &g7 5.892 0-0 6.2ge2 e5 7.h3 Qe6 8.2e3 exd4 9.2xd4 HeB 10.0-0 2d7. In our move order, White may consider saving the time spent on 9.h3, but in view of the natural development with £c1-e3 (h2-h3 makes ...g4 impos- sible) he rarely gains anything with this. Much more drastic than those subtleties is 8.2g5!; after 8...297, 9.05 poses some unpleasant questions, Black prob- ably has to play the subdued 8...2.e7. Therefore in Feistenauer-Hadjieff, Austria 1997, Black chose 7...2e7. The continuation was 8.e4 0-0 9.Ee1 He8 10.b3 We8 11.2b2 294 12.Wd2 Ded 13.14 Qg6 14. Od7 15.Eae1 c6 16.Wd3 28 17.Qd1 We7 18.c4 Hads 19.e3 £c8B 20.Adf5 Ad7 21.We3+ etc. As indicated above (see note a3 to Black's 4th move), the ‘&e7-Pirc’ can rarely be considered a_ satisfactory approach for solving Black's opening problems. The consequences are typical: Black fails to find a constructive plan, shuffles his pieces around and therefore worsens his position even more. The following game is mentioned mainly because of its final position: 5...207 6.292 2\xd4 7.Wxd4 G46 8.b3 0-0 9.2b2 c5 10.Wd3 Sb8 11.0-0-0!7 b5 12.e4 We7? 13.axb5 Wa5 14.2\xd6! Bab (14.208 15.04!, 14...Wxa2!? 15. e5 94 16.415 2xf5 17.Wxf5 Schlenker) 15.Wce3 Wxa2 16.415 2d8 17.2xd8! Ebxd8 18.2e7+ Wh8 19.Wxf6! and 1-0 because of 19...gxf6 20.2.xf6#, Schlenker-Mirl, Tettnang (rapid) 1994. 6.292 g616 Chapter | The second player imagines the problems after the passive 6...2f6 (= 5....26) and therefore tries his luck with some kind of ‘counterattack’ (...2.97) against the d4- knight. 7.51? The trademark knight tactics. However, after 7.Aaxc6!+ Black’s situation be- comes critical immediately: a) 7...&xe6 8.2xc6+ bxc6 9.Wd4 “6? (9...WI6 10.Wa4+) 10.295 2e7 11.04 b) 7..bxc6 8.Wd4 Wie 9.Wed+ (9. Wo4!?AWb7) 9...We7 (or 9...2e7) 10.\b5!. Therefore, Black is forced to make serious concessions like 9...%2d8 (forfeiting castling) or 8...f6. 7...2.97 8.2b5 Hc8 9.2e3 Played just to prevent ...a6, which would force a retreat. 9...a6? Acting as if there were no danger at all. a) 9..Age7?! 10.295+ (10.2dxc7+ Eixc7 11.Wxd6 a Schlenker). b) 9...8xb2 10.2lb1 with more than sufficient compensation. c) 9...2f6!, and Schlenker’s suggestion 10.2\dxc7+ doesn't quite manage to break through: 10...2ixc7 11.2xd6+ &f8 12.@\xb7 Exb7 13.2xc6 Exb2. In the further course of the game, Jérg Pape (a strong player who is much feared at local quickplay events) gets crushed: 10.Abxc7+ &f8 10...2x07 11,2b6+-. 11.2b6 Wg5 12.0-0 Hf6 13.¢e3 We5 14.214 Wxb2 15.2b1 Wd4 15...Wxa2 16.2xf6 2xf6 17.Wxd6+ +-. 16.2xd6+ &g8 17.Wxd4 Dxd4 18.2\e7+ And we have a windmill combination. 1-0 Section 2 1.268 e5 2.513 205 (2...05) Game 2 Mehihorn — Vitkevicius corr, 1992 1.2¢3 €5 2.013 2.05?! An idea which stems from some gambit circles. From an unbiased point of view, it is very doubtful despite some pretty variations. The idea - exchanging the 5-pawn for the f2-pawn, simultaneously preventing White from castling - is well- known and mostly unfavourable for the party that drops the e-pawn. For instance, 1.64 e5 2.43 Ac6 3.23 ATE 4.4047! ®xe4! 5.2x17+? (5.0-0!) 5...2xt7 6.2xe4 d5+. Or, with colours reversed, 1.e4 e5 2.43 Deb 3.203 205? 4.Dxe5! &xf2+ 5.Sxf2 DxeS 6.d4 Ac6 7.23 d6 8.22 6 9.8f1+, Capablanca-Liebenstein, New York 1913. This is basic knowledge, but | doubt that gambit folks are always aware of this. | don't appreciate other ‘gambit’ moves like 2...216?, 2...15? or 2...e4? (all of them have occasionally been tried) at all, with the single exception of 2...d5. After all, it1.c3 e5 17 is problematic to reject the offer as 3.e4?! d4 4.2 leads to the inferior line 1.23 d5 2.e4 d4 3.4\ce2 e5 4.437! (see Game 17). The idea may be illustrated by the following example: 3.2xe5 d4 4.4\b1 Wd5 5.23 216 6.d3 Dc6 7.04 dxe3 8.2.xe3 £94 9.262 0-0-0 3 2b4 11.0-0 &xc3 12.bxc3 Bhe8 13.Qd4 2xe2 14.5xe2 Dg4 15.202 15, Balzer-Lind6rfer, Simmersfeld 1996. In the meantime, Black obtained useful compensation. It is all too easy after the event to offer some suggestions for an improvement (maybe 6.e3) - but when playing such a position over the board it is very easy to miss the point at which one should give the right direction to the game. 3.2xe5 A poor move is 3.e4? 2b6!F, 4.axe5? fails to 4...We7, and after other moves Black will hunt White's knights. Aasum gives 3...We7 (instead of 3...2b6) a ‘!’, a strange judgement in view of 4.2\xc5 Wxc5 5.d4+ because of the bishop pair. If one likes to reject the gambit offer, then 3.€3(!) Dc6/d6 4.d4+ suggests itself. This may on its own be sufficient to prove the uselessness of the move 2...2c5. 3..2xf2+ 4cexf2 Whae 5.cogt!? 5.213? Wi6+ 6.ve4 d5+ or 5.ve3 Wg5+ 6.chd4 d6 leads to spectacular king-hunts. It may be fun to analyse those positions, but it is meaningless. It's obvious that White should not go in for this, as he has some risk-free and promising alternatives. Among those, 5.g3 should be men- tioned. However, White should not underestimate matters, the weakened squares g4 and h3 may later be used for some tricks. For instance, 5...4/d4+ 6.e3 (also good is 6.%6g2 Wxe5 7.d4, Mestrovic-Jakubovic, Bihac 1999) 6...Wxe5 a) 7.d4 Wi5+ 8.ceg2 (8.137 Wxc2+, 8.491 would be a concession as well) ...d5 (Aasum-Roscher, corr. 1992) fol- lowed by ...2\f6 or 9.203 Wh3+. b) 7.Wf3 AMG (7...c5!?) 8.d4 We7 9.e4 d6 10.h3 0-0 11.2c4 Dc6 12.263, Laird-Bullockus, corr. 1980, 12...Be8 13.a3. Black may consider 12...2d7 followed by ...2ae8 or 12...266. c) 7.292. Probably, White also keeps an advan- tage in those cases, but at least Black is not without counter-play here. 5.%g1 offers no target for an attack at all, and therefore White may pursue his main goal undisturbed, namely the building up of a dominating central position. He can bring the h1-rook into play later, as Black's development isn't terrifying at all. 5...Wd4+ 6.e3 Wxe5 7.d4 Wa5?! 8.e4+ d6? 9.b4!+—18 Chapter | A fair punishment for an amateurish approach - moving around with the queen. 9...Wxb4 Or 9...Wb6 10.d5 We6 11.s2b5!. 10.205 2.94 (10...Wa5 11.8d2). At the same time resigning due to 11.Wxg4 Wxd4+ 12. £263, threatening a mate on c8. 1-0 Section 3 1.2c3 e5 2.0f3 Acé 3.d4 exd4 4.4)xd4 various Game 3 van Geet - VI. Sokolov Wijk aan Zee 1970 1.Ac3 e5 2.513 2c 3.d4 exd4 4.0xd4 This position will accompany us through- out the next sections. Very different moves had been played at this point, but the only unflawed ones are 4...205 and 4...2f6 (the same moves that are considered best after 1.e4 e5 2.213 Ac6 3.d4 exd4 4.xd4!). But only in exceptional cases does White go for a transposition to the Scotch by €2-e4. On the agenda are the manoeuvre ©xc6 followed by Wd4, knight tactics with @db5, Qd5 and/or Df5 as well as the fianchetto of the king's bishop with g3/2g2. 4...96?! Others: 4...06?! > 1.203 e5 2.23 d6 3.d4 exd4 4.2\xd4 @\c6, Section 1. 4...816? (or 4...Wh4? 5.Adb5), played in several (correspondence) games, can only be explained by inappropriate think- ing of analogies to the Scotch Game. Black is already losing, e.g. 5.2db5 Wa8 (or 5...2c5 6.2xc7+ Sd8 7.e3) 6.2.4 d6 7.25 1-0, Tilstra-Faber, Leeu- warden 1992. The rarely played 4...2ge7?! seems, at least, not to run into a direct refutation. 5.e4 Axdé (5...96? 6.295+- 297 7.Dd5 Bxd4 8.Yxd4!) 6.Yxd4 Ac6 7.Wed is a rather good version of the Scotch, 5.295 is also unpleasant to meet. Usually, the move 4...d5, is categorised among the unfavourable minor variations as well. In my opinion, 4..d5 is better than its reputation; sensible play by both parties should lead to a normal opening initiative for White, but nothing more. And this is already an achievement for White, compared to, e. g. 1.d4 cS 2.23?! cxd4 3.xd4, and after the correct reply 3...d5! (threatening ...e5 - but not 3...e5? 4.2\b5 d5? 5.Wxd5) Black already has the advantage - it's called ‘¥' by the authorities. Look at 4.25 a6 5.Wxd5? axb5. Unlike in this demonstration line, in our case the typical Left Knight's motifs are helpful: the Wd4 move, the possibilities of an accident on c7, etc.1.2)¢3 e5 19 a) 5.Axc6 Also possible is 5.e4, transposing to 5.Axc6 after 5...52b4 6.Axc6 bxc6 (this was actually played in van Geet-Sanders), besides, Black may try to fight for equali- ty with 5...dxe4 or 5...2\xd4 6.Wxd4 dxe4 7.Wxd8+ Sxd8. 5...bxc6 6.e4 The resulting positions are very similar to the Scotch Four Knights. 6...2.b4?! 6...d4 (Aasum) 7.Ae2 5 8.c3 with an advantage. But preferable is 6...f6. Unlike in the Scotch Four Knights, where Black plays ..d5 only in answer to 2d3 (1.4 e5 2. NB Acb 3.003 D6 4.04 cxd4 5. Axd4 2b4 6.Axc6 bxc6 7.2d3 d5 8.exd5 cxd5 9.0-0 0-0 10.295 etc.), here White may deliver a check on b§ after 7.exd5 cxd5 8.2b5+; however, after 8...2d7 9.0-0 £e7 10.295 c6 11.2d3 h6 12. 2x6 Oxf6 13.5e1+ S18 14.WI3 g6 15.\e2 “g7 he didn’t achieve anything in the game Santo Roman-Barbero, Luzern (Wch teams) 1985 (with a some- what different move order). 7.4! The key move, which will be seen repeatedly in the sections which follow. 7.exd5 &xc3+ 8.bxc3 Wxd5 9.Wxd5 cxd5=, H.Maier-Schénlein, Erlangen 1994. 7...2x3+ 8.Wxc3 WE 9.2d2 Wxe3 10.2.x03 6 10...{6 cannot make Black's disadvantage disappear either (bishop pair, White's better pawn structure). 11.2xf6 gxf6 12.exd5 cxd5 13.2b5+ 207 14.2xd7+ &xd7 15.0-0-0 Scé6 I'm not sure whether the position really is +, as van Geet says - strong defence may save a draw. However, after 16. Ghe1 Bae8 17.oxe8 Uxe8 18.d7d2 van Geet finally won the pawn ending, van Geet-Sanders, Rotterdam 1953. And the winner is always right... Nevertheless, this game shows, as well as many others, that 1.2\c3 isn’t at all an opening exclusively for players who like chaotic positions. Frequently, good technique is called for - 7.#d4! is a strong ‘technical’ move, subsequently followed by a simplification into a technical rook ending. b) 5.24, eyeing up an assault on c7, e.g. after 5...06?? 6.2db5 (5...2057? 6.Adb5) or 5...2d6?? 6.2xd6 Wxd6 7. Qdbs Wd8 8.2\xd5, each one leading to a disaster. Both sequences really have happened, in correspondence chess and in a high level German team competition. Also second-rate is 5...2e6 (one plays this reluctantly at best) 6.Acb5 2b4+ (6...21c8? 7.2xc6 bxc6 8.2\xa7) 7.c3 a5 8.Axc6 bxc6 9.24 with a nice advantage, Hendriks-van Ruitenburg, Leiden 1999.20 Chapter | After 5...2xd4 6.Wxd4 c6 7.e4, Black escapes into some kind of -position by 7...dxe4 8.Yxd8+ &xd8 9.4\xe4, Semrl- Srebrnic, Ljubljana 1999, or 7...26 B.exdd5 Dxdd5 9.axd5 Wxdd 10.Wxd5 cxd5, Hermlin-Tikander, Lahti 1997. 5...2.b4! 6.axc6! After 6.e4?, the response 6...Wf6! is surprisingly unpleasant, Brauning- Kretschel, 2nd Bundesliga 1986. 6...2xC3+ 6...oxc6?, and 7.WWd4 already wins a pawn (7...2x03+ 8.Wxc3, xc6, g7), un- less Black replies 7...s2f8. T.bxc3 bxc6 8.d! + according to Brauning/Moser. Black should be able to equalize after 8...2f6 9.04 0-0 10.295 (K.Wilhelm-Raijmaekers, corr. 1994) 10....£e6 - in my opinion, the lack of development hinders White from playing for an advantage. Back to 4...96: 5.25! 5.2xc6 bxc6 6.Wdé is parried by 6...\4f6! (Bent Larsen in Kaissiber 17). Perhaps White has chances for a slight advantage. We are experiencing our first lesson in ‘knight tactics’ with 1.2\c3 - many more will follow. 5...a6 5...297 6.2b5 2e5 is unfavourable. The only question is in which manner White will clarify his advantage. a) 7.4 a6 8.fxe5 axb5 9.2.95 Wxg5 10. Dxc7+ #d8 11.)xa8 Wh4+ 12.93 Wb4+ (12...Wa4 13.Wd6!, picking up the h8- rook after 13...Wxa8) 13.Wd2 Wxb2 14. Ed1+. Probably the complications after 9.295 are unnecessary, 9.4 is suf- ficient to prove an advantage. b) 7.214 2xi4 8.axf4 Qge7 9.h4, and this is + or more. In Hermlin-Luukkonen, Jyvaskyla 1996, Black bought his freedom with 9...d5!? 10.Axd5 &\xd5_ 11.Wxd5 We7 12.0-0-0 0-0. After all, White's extra pawn is not so crucial here. 5...2)b4 is Timman’s suggestion to limit the damage. After 6.0b5 @\xd5 7.Wxd5 Black has to deal with dangers on c7 and a7(!): We5+ is threatening, 7...2g7? 8.We5! or 7...d6 8.2xa7! (OK, this one was unearthed by a computer ...). Black can maintain the material balance only with the ugly move 7...f6. 6.295 Black has better chances of survival after 6.Axc6 bxc6 7.Wd4 16 8.Wed+ wf7, Pel-Kotan, Piestany 2004. 6...f6 6...Wxg5 7.2\xc7+ d8, and at this point Bent Larsen (in Kaissiber 17) analyses 8.2xa8 to some extent. However, 8. &\(any)e6+ wins the queen - as of course was already mentioned by van Geet. 7.24 297 8.e4 DHE 9.204 Dad?! 10.e5!1.23 e5 21 10...95 Admits defeat. On 10...A\xc4 there fol- lows 11.exf6 7 (11..0-0 12.Ae7+!) 12. ENB He8+ 13.61 2h8 14.2\g5+ wg8 15. De7+ Bxe7 16.fxe7 Wxe7 17.Wd5++-. 11.exf6 2xf6 12.Wh5+ o18 13.Wxh6+ 297 14.Wxg5 Letting Black go into an ending a pawn down. In similar positions, one is tempted to sweep the opponent off the board. But a bird in the hand is the safer choice, especially since White retains a positional advantage as well as his extra pawn. 14..Wxg5 15.2xg5 Axc4 16.0-0-0 c5 17.2e7+ &f7 18.2xc5 And Black dragged on from hopeless position until move 64. we 10 this Section 4 1.203 e5 2.2\f3 Dc6 3.d4 exd4 4.\xd4 \xd45.xd4 About ‘non-moves’ and ‘non-theory’ Really a ‘non-line’. The area of the Scotch Game and similar open games has been reached; the databases are fairly full of games, but nevertheless one doesn't find anything about this in the literature (even not in the ancient books by Pachman or Keres), or at best some off-hand notes in the Encyclopedia. However, the degree of stupidity (1.64 25 2.543 Dc6 3.04 exd4 4.axd4 Dxd4? shows a certain amount of ignorance) does not always correspond to the damage it causes: If Black continues perfectly, the Scotch ‘non-line’ is not even a definite +. Game 4 Frick — St. Nikolic Tubingen (invitational) 2001 1.Ac3 e5 2.13 Dc6 3.d4 The thematic move. Of course, 3.e4 leads to the Three or Four Knights Game, and after 3.93, some kind of Pirc Defence with colours reversed arises, where the move 4\b1- c3 does not really add its own stamp to the game, e.g. 3...d5 4.d3 S46 (or 4...f6 5.292 2e6 6.0-0 Wd7 7.e4 d4 8.45, Schlenker-Wo.Haist, 1984) 5.292 2e7 6.0-0 0-0 7.295 2e6 8.e4 d4 9.22 etc., R.Rodriguez-Handoko, Jakarta 1993. 3...f5!? (instead of 3...d5) seems interesting to me, in order to present no target for the c3-knight. In my opinion, if White wants to go for a Pirc reversed, he would do better to open the game with 1.93. (compare also Sections 41 and 42) After 3.e3 d5, 4.2b5 might be an idea (© Section 36), but 3...f5 looks good. 3...exd4 3...d6 © 2...d6 3.d4 c6 (Section 1). The ‘generous’ 3...e4?! was played in probably the oldest (according to Adrian Harvey in Kaissiber 15) authentic 1.2c3 game: 4.2xe4 d5 5.203 215 6.e3 6 7.203 Re4 8.2xe4 dxe4 9.2195 We7 10.0-0 h6 11.4h3 g5 12.44 g4 13.42 hS 14.WWe2 0-0-0 15.a3 Hg8 16.b4 g3 17.hxg3 Exg3 18.b5 Axd4 19.exd4 Exc3 20.2b2 Hg3 21.c4 Ag4 22.2 x94 hxg4 23.201 Wh4 24.263 2g7 25.212 Wh8 26.2xg3 2xd4+ 0-1 (perhaps resigning a bit early on account of 27. 2f2 g3 28.Wg4+, Bicker), Kieseritzky- Desloges, France 1847. 3...4e7? already costs a pawn: 4.4\d5 Wd6 5.dxe5 @\xe5 6.2)xe5 c6 (6...Wxe5? 7.5214) 7.217 (7.21412 Wxd5 8.Wxd5 +) 7...&xf7 8.23, Hohm-P. Schmidt, corr.22 Chapter | 1982. After 8...Wxdi+ 9.axd1 d5, Black hardly has any compensation. 3...f67! is, of course, a second-rate move. In such cases White should return to a classical approach with 4.e4. 4.2xd4 Dxd4 If you ask a master to make a list of can- didate moves in this position, you may find...2¢5, ...2f6, ...d5, ...sb4, then perhaps ...Wf6 or ...g6 on the said list, because all of these moves represent one idea or another. As a second step, he will examine the concrete risks or disadvantages of the several possibilities, and the list becomes smaller. 4...2\xd4 should not even make it into the preliminary list, because there are no objective reasons which speak for it. At best, emotional reasons will dictate this choice: one feels uncomfortable because one doesn't know ‘theory’, one is worried about some concrete pos- sibilities in the main lines, finally the subconscious desire for simplification takes over. And actually, according to my working database one out of seven players chooses 4...\xd4 at this point. The Scotch ‘non-line’ 1.64 e5 2.43 2c6 3.d4 exd4 4.2xd4 2xd4 is the choice in one case out of 17, therefore it's played a bit more often than 4...W4h4, to which move GM Lev Gutman dedicated a whole book... 5.xd4 Theory sets great store by the advantage of a centralised queen, which cannot be driven away by natural moves. So, in the case of 1.e4 e5 2.43 Ac6 3.2b5 dé 4.d4 exd4 5.Wxd4 2d7 6.2xc6 &xc6 White must pay for this privilege by part- ing with his bishop pair. In a seemingly similar line of the Philidor Defence - 1.64 e5 2.213 d6 3.04 exd4 4.Uxd4 ATE 5.295 &e7 6.4c3 - Black, on the contrary, has ...2)c6 at his disposal. 5.16 This is much more often played than 5...Wi6 which is treated by van Geet. Almost every 1.2\c3 devotee knows the game van Geet-Taksrud with its startling rook manoeuvre, but from a theoretical point of view this game is less relevant: 5...Wi6 6.Wad 6.Wxf6 Axf6 7.Db5 2b4+=. 6...06 6...Qe7 7.2Qb5 Web 8.2f4 Add 9.We4++ 2e7? 10,Wxd5 +-. 6.205 7.5e4£ bishop pair, Pommerel- Robidoux, corr. 1997. In the case of 6...%c6, White can play 7.Wxc6 (— because of the doubled pawns) or try 7.2\b5!? (van Geet). 7.N4!? h6 7... 2.87 8.2.95 Wed 9.214 WS 10.93 d5 11.23, Glowatzky-Peterle, corr. 2004, 11... WIG 12..8.xc8 Hxc8 13.0-0-0+. 8.5h3 d6 8...De7 9.513 b5 10.Wa5!? (or 10.2xf6 with an endgame edge) was played in the above mentioned game van Geet-Taksrud, corr. 1982/83. On his CD, van Geet him- self offers the following nice game: 9.295 WIS 10.He3+ 2e6 11.94 Wg6 12,245 hxg5 13.h5 Hxh5 14.gxh5 Wxhs 15.Ac7+ &d8 16.2xa8 1-0, Montsma- Kemme, Netherlands 1998. 6.e4 After ...2\f6, this classical move is logical, because otherwise White would have to reckon with ...d7-d5. Pure piece1.23 e5 23 play seems less appropriate here, e.g. 6. 2g5 2e7 7.2e4 Dred 8.¥xe4 d5!= 9. Sxe7 Wxe7 10.Wxd5 Wb4+. However, if Black chooses 5...d6, then White - be- sides 6.e4 - may as well go for a flank- opening setup by 6.93 or 6.3. An odd idea is 6.Ab5 from some U-14 tournament, seemingly with the double- threat We5+ and “xa7 - time and again computers show us how often such moves have gone unnoticed. Here, however, this idea fails due to 6...2€7 7.2\xa7? cB. 6...d6 But not 6...s267? 7.e5!+. This position can also occur after 1.e4 5 2.213 Dc6 3.d4 exd4 4.Axd4 Dxd4 5. Wxd4 d6 6.203 Af6 or 1.e4 e5 2. QB Acb 3.3 D6 4.d4 exd4 5.2\xd4 &xd4 6.WWxd4 d6. However, in the whole theory of the Scotch one cannot find a reasoned opinion about this position. In his book Grundlagen des modernen Eréffnungsspiels, A.Suetin mentions that after 1.64 @5 2.213 Z\c6 3.4 exd4 4.0xd4 Dxd4 5.Wxd4, on 5...216 as well as 5... WH there is the strong reply 6.e5!, and further shows an old miniature with 5...2\e7 6.204 c6. He does not mention 5...d6 which is best, restricting the e4-pawn. The Encyclopedia gives 1.e4 e5 2.03 Deb 3.d4 exd4 4.xd4 Dxd4 5.Wxd4 d6 6.Ac3 G6 7.8.95 27 8.0-0-0 with the optimistic assessment ‘+’, no further information. Dunnington quotes a game which shows lousy play by Black. As men- tioned, all other sources remain silent. 7.2.04 The databases display a multitude of sensible moves like 7.204, 7.295, 7. S2e2, 7.214 etc, each scoring about 65- - 70% for White. But still, all those games do not reveal a forcing way to an edge. In such cases, opening authors tend to be over-careful on their readers’ behalf. From the multitude of games, they pick a drastic example, skimming over later inconspicuous inaccuracies - after all, it can serve as ‘visual instruction’. Here, A.Dunnington in Winning Unorthodox Openings: 7.214 26 (?, beyond doubt, 7...%@7 8.0-0-0 0-0 is more sensible - H.K.) 8.0-0-0 £.e7 9.e5! dxe5 10. Wxe5 2d6 11.8b5+ d7?_—(11...8f8 + Dunnington) 12.Exd6! cxd6 13.\/xg7+-, Demuth-Michels, St.Ingbert 1991. Then, over the board, one usually has the following experience: one remembers the fact that "there is” a clear-cut refutation. In one's memory, the details remain vague, but still - one should be able to figure out such a simple line over the board, shouldn't one? Then, while calculating several pos- sibilities, unexpected resources for the defender emerge, while simultaneously quiet lines which promise only a slight advantage are discarded. No matter which decision one makes - one starts the battle with a feeling of confusion about a missed opportunity. After coming back home, one looks up, reasoning: ‘Oh well, if my opponent had played so poorly as the guy in the book, I'd certainly have swept him off the board!” Usually, the reality on the chessboard is much more complex (and possible ‘re- futations’ much less clear) than so-called “popular” books want to make us believe. Now back to the alternatives to 7.204, e.g. with the idea of castling queenside: 7.295 2e7 8.0-0-0 (+ Encyclopedia, as mentioned) 8...0-24 Chapter | Now White achieves nothing by playing 9.e5: 9...dxe5! 10.Wxe5 2d6 11.Wd4 2e6 12.4\e4 2e7=, Mahmoud Abdelkhalek- Langenfeld, Germany 1996. Therefore, he must try to play for a kingside attack - and at least optically his chances in a struggle on both wings are looking better. However, if Black aims for simplifications he should hold his own: 9.f3 Alternatives could be 9.2.4 or 9.f4. 9...2.06 10.3 Or 10.94 4d7! 11.s2xe7 (11.2e3 216) 11...Wxe7 12.h4 Wi6! 13.We3 Ded 14.202 Agé6 15.b1 Wt4 16.Wd4 Wed with good equality, R.Schmidt-Stritzel, Pinneberg 1998 10...06 11.204 Wa5 12.2d2 Wb6 13. Wxb6 axb6 14.2xe6 fxe6, Koc-Socko, Koszalin 1999. Subsequently, the Polish GM Socko scored an instructive win thanks to his centralized pawn mass (15.295 G7 16.%eb1 e5 17.e2 Ye6 18.2xf6 gxf6 19.c4 Ha5 20.3 15 21.Ehf1 Haa8 22.c2 Efé 23.43 Lats 24.see2 Eh6 25.0h1 Eg6 26.Ehg! Bigs 27.93 h5 28.292 2d8 29.0dg1 h4 30.94 fxed 31.fxe4 b5 32.cxb5 b6 33.Hf1 Exg4 34.5xg4 Exg4 35.bxc6 bxc6 36.54? Exe4+ 0-1) Back to the main text. The concrete moves 7.2)d5 (leading to a rather harmless line of the Belgrade Gambit: 1.e4 e5 2.43 26 3.203 AE 4.d4 exd4 5.25 dé 6. Dxd4 Dxd4 7.Wxd4) and 7.€5 (7...dxe5 =, 7...2g4=) are not fruitful. Black's comparatively favourable ex- periences with this line are consistent with the principle that in positions with a space disadvantage (d6-pawn against e4-pawn) every exchange of a minor piece benefits the defender (as stated repeatedly by Dvoretsky). With some heretical exaggeration one could apply a ‘I’ to the Scotch non-move (1.e4 e5 2.4N3 Ac6 3.d4 exd4 4.Axd4) 4...axd4, adding the following explanation: “After 1.e4 e5 2.43 @c6, the central advance 3.d4 is fruit- less, as Black - unlike in the Philidor Defence 1.e4 e5 2.013 O6 3.04 - can immediately ex- change a pair of minor pieces’. Back to 7.2.04: 7..2.eT 8.0-0 0-0 The most widespread model of opening theory is the tree structure - on each level, there are more branches and ramifications. But occasionally it is vice versa, like a river which widens as more and more tributary streams flow into it. So our river, starting with 1.203 and a proper Scotch, now includes games from all kind of Open Games, e.g. 1.e4 e5 2.493 cb 3.204 (or 3.d4 exd4 4.204 6, Balzar-Hort) 3...Af6 (or 3...207 4.44 exd4 5.Aaxd4 D6 6.213 @xd4 7.Wxd4 d6 8.0-0 0-0) 4.d4 exd4 5.0-0 d6 6.2xd4 Axd4 7.Wxd4 2e7 8. 4)c3 0-0, as already played in Teichmann- Mieses, London 1895. 9.295 As on the 7th move, the choice is a matter of taste. Alternatives are 9.h3, 9.24 or 9.Be1 9...c6 And now, 9...2e6 or 9...h6 were also possible - in the grey area between equality and slight advantage for White.1.203 e5 25 10.a4 It is not by chance that | have picked this very game Frick-Stanimir Nikolic from the many candidates. The first player is a well-known 1.4\c3 expert, and the black pieces are not played by a N.N. but a grandmaster who is known for his lack of prejudice (malicious gos- sip is that he knows less opening theory than the average amateur, but this does not get to the heart of the matter). For instance, St. Nikolic always battles against the Sicilian with 2.2c4, and it is only recently that others have taken up this idea (e.g. no less than Adams by playing 1.e4 05 2.613 d6 3.24). Not long before, there was another de- cidedly pragmatic grandmaster playing this position with Black: 10.2fe1 hé 11. 2h4 Wh6 12.Wd3 Wed 2-'2, A. Balzar- Hort, Bundesliga 2000. 10...Wa5 With 10..d5 11.exd5 Dxd5 12.2xe7 ®xe7 13.Had1 Wb6, Black comes close to equality, Ljubic-Tokovic, Hradec Kralove 1992, just as after 10...Wb6. 11.24 Ed8 This looks somewhat unnatural to me (weak point f7). After 11...We5 or 11...2e6, White at most has this indescribable nothingness which is the advantage of the first move. 12.2fe1 Wh5 Due to the possibility of x17, White now plays energetically: 13.e5! dxe5 14.xe5 Wxe5 14..Wxh4 15.axf7+!. 15.Exe5 2d6 16.0e3 2f5 17.243! 2xd3? Much better was 17....2g6! (nearly =). After 18.2xf6 gxf6, the bishop holds together the kingside, and Black can point to his pair of bishops - as long as White doesn’t straighten out Black’s pawn structure by 2d3xg6. In the text, we will soon see a thematic position with a good knight against a bad bishop. 18.2xf6 gxf6 19.Exd3 2e5 20.Bad1 Exd3 21.2xd3 Se8 22.ef1 Be7 23.ad1 23...f5? Better was 23...h5. According to Zeller in the tournament book, this was still only +. 24.2)e3! 6 Because of 24..f4? 25.5d8+ 26.215 +. 25.2xf5 c7 26.c3 The technical phase which follows was handled with an astonishingly sure touch by Christoph Frick against the grand- master: 26...247 27.93 Seb 28.De3 Eg7 29.14 2c7 30.ve2 Eg8 31.03 Eb8 32.94 a5 33.05 h5 34.8e3+ wd5 35.gxh5 b5 36.h6 bxa4 37.h7 Eh8 38.5e7 £d6 39.Eb7 a3 40.bxa3 2xa3 41.2)h4! a4 42.2\g6 Exh7 43.2xh7 &b2 44.27 a3 45.h4 1-0 &g726 Chapter | Section 5 1.2c3 e5 2.23 Ac6 3.d4 exd4 4.2\xd4 a6 5.2.95 This is a variation which occurs rather often and offers excellent chances for an early knockout. | am amazed that on his CD van Geet covers this topic in only 17 lines. On the other hand, a quite comprehensive in- troduction can be found in Dunnington's Winning Unorthodox Openings. In fact, there are many games and much analysis available, and in this section | have rather compressed the material. Also, there are a lot of possible things which can go wrong for Black. Among the many plausible moves, only two (5... 2b4 and 5...£c5) do not immediately ruin Black's game. The inferior lines will be covered in Game 5, the viable ones in Game 6. Game 5 Frick — M. Strobel Germany 1994 1.2463 e5 2.0f3 DAc6 3.d4 exd4 4.2\xd4 Af6 5.8.95 The most rigorous continuation. 5.e4 is a Scotch Four Knights, as well as 5.2)xc6 bxc6 6.e4 (6.93 from Malm- strém-Sorroche, corr. 2001, is nothing special) 6...22b4. But in this move order, 6...d5 seems to promise easy equality (© 1.04 e5 2.013 Ac6 3.d4 exd4 4.dxd4 M6 5.2xc6 bxc6 6.4\c3?! instead of 6.e5). 5.444 (5...2b4, 5...205) is playable, but far less energetic than 5.295. The real alternative to 5.2.95 is 5.93: a) Inthe 1.4c3 literature, this possibility is usually mentioned in the same breath as the game Brauning-Rapp, 2nd Bundes- liga 1985. There followed 5...2.¢5 6.b3 £e7? (6...2b4 7.8.92 0-0 8.0-0 2xc3 9. bxc3 d6 10.8b1 He8, Brauning-Kobese, Germany 1993. White has some initiative, e.g. after 11.Qd4) 7.292 0-0 8.0-0 dé 9.e4 He8 10.h3 Hb8?!. As we already know (see Game 1), Black has a lousy version of a g3-Pirc. The next moves show Black's fruitless attempts to regroup his minor pieces, 11.f4 £18 12.263 a6 13.Wd2 Ad7 14.Badt Se7 15.2a7 Ea8 16.212 Dgé 17.Bfet c6 18.204 6 19. 94, and soon he was crushed. In my opinion, the game doesn't prove much besides the fact that Black lacked the necessary feeling for the position. (compare Feistenauer-Hadjieff_ within Game 1, Black's 5th move deviations) b) The logical answer is 5...d5 6.2.92, as played in two games by Ekebjzerg b1) 6...xd4 7.Wxd4 2e6 8.0-0 @ 295 £e7 9.0-0-0 0-0 Vitomskis) 8...e7 9. Ed1 c6 10.e4 dxe4 11.We5 Wb8 12. Wxb8+ Exb8 13.2\xe4 Dxe4 14.2xe4 0-0 15.2e3 a5! 16.%bg2 Efd8 17.2b6 Exd1 18.Exd1 a4, and Black was able to hold his own, Ekebjeerg-Vitomskis, corr. (Weh Y-tinale) 1989-1993. b2) 6...b4 7.2xc6 bxc6 8.0-0 0-0 9.Wd4 c5 10.Wh4 2xc3 11.bxc3 Ded 12.Wh5 £5 13.c4 c6 14.cxd5 cxd5 15.c4 2a6 16.cxd5 Wxd5 17.263 Hac8~, Ekebjzerg-Steiner, corr. 1987-1992. Now back to 5.2.95:1.23 e5 27 5.2.07? Or: 5.6? This ‘routine question’ to the g5-bishop more or less costs the game, as do the stereotyped moves 5...d6? or 5...2e7?. It's remarkable that recently a modern 2400 player fell into it, 5...h6? 6.24?! d6?! 7.2x16 gxf6 8.Ad5 2e6 9.4 Axd4 10.Wxd4+, St.Berger-Sarakauskas, Ham- burg 2003. 6.2x16! Wxi6 7.adb5 The prelude to a nice little exercise in knight tactics 7...We5 Comparatively the most demanding. Other moves: 7....3.b4 8.2xc7+ followed by 9.4)7d5. 7...%d8, and now of course not 8.Ad5, as given by van Geet on his CD (8...Wxb2!e), but 8.Axc7!+- (S. Ullrich-E.Rotshtein, Germany 2004). 7...2,05!? is a bluff: after 8.2e4 Black has a way out with 8...Wf4! (Bellon-Hess, Dortmund 1980), and after 8.axc7+ #8 9.2xa8?? 2xf2+ 10.2 Wi4+ White gets mated. But instead of 9.xa8, White just has to play 9.e3!+-. The retreat 7...Wd8 is very passive. After 8.Ad5 2b4+! (8...2d6 9.Dxd6+ cxd6 10.93, strategically winning) 9.c3 2a5 (Twardowski-Dworakowski, Koszalin 1998) or 9.xb4 (Dunnington) 9...2\xb4 10.Wd2 Ac6 11.We3+ HB (11...0e7? 12.Wc3) 12.0-0-0, Black can limit the damage (barely + in both cases). 8.205 8.147! Wxf4 9.0d5 Wh4+ 10.93 Wed on Motwani, but easiest is 8.\Wd5!+ (8...d8, 8....2d6). bs 8...206 9.4! +-, 8...Wxb2 9.Hb1! (but not 9.Qbxc7+ dB 10.Axa8? Lb4+ 11,.2xb4 We3+! 12.Wd2 Wxat+ 13.Wd1 Wxd1+ 14.exd1 @xb4-+) 9...Wxa2 10.Abxc7+ &d8 11.2xa8 +- Motwani. 9.bxc7+ &d8 In fact, this is still a bit tricky. In the game H.Steiner-Morris, corr. 1995, 10.@b5 (probably not the best) 10...Wxb2 was played, and Black finally managed to get a draw. 5...d6?. After this things are even easier compared to 5...h6, as after 6.2.xf6 Wxté (6...gxf6 +/+-) 7.Qdb5 Black is deprived of his swindling chances connected with ...WeS, ....2b4(+) or ...2d6. In the game Bellon-Medina, Spain 1979, Black resigned after 7...%08 (7...Wd8 8.2)d5) 8.2\xc7!. 5...d5?! 6.e4 (one may compare, with colours reversed, 1.e4 46 2.2\c3 d5 3.exd5 Axd5 4.44 DAc6 5.3 £94, and, for argument's sake, let Black be allowed to make another move simultaneously: 5) 6.807 7,505 207 8.exd5 Dxd5 .DxdS 2xg5 10.We2+ De7 (10...2e7 11.0-0-0+, 10...%f8+) 11.We5! 2xb5?! 12.Axc7+ &f8 13.Qde6+! 1-0, Dunst- Gresser, USA 1950/51. Somewhat better is 6...2b4, whereafter Black narrowly survives after 7.Axc6 &xc3+ (7...6xc6? 8.Wd4 with an edge, Caruso- Sbarra, Verona 2005) 8.bxc3 bxc6 9.e5 (9.exd5!?, hoping for the bishop pair) 9...We7! 10.We2 h6!. Instead, 7.2b5 should give some advantage. 5...21ad4?! 6.Wxd4, e.g. 6...2e7 7.4 dé => Game 4, 7.295. 6.215!+ 0-028 Chapter | Moves such as 6...2f8 or 6...2g8 had been played as well, obviously born out of desperation. Probably the best try is 6...h6! (6...d6 7.Axg7+ £f8 8.Wd2!+-, van der Spek-van der Plas, Netherlands 1992; 6.5 7.axe7 Dxe7 8.e4+-, Hendriks-van der Beek, Hengelo 1998) T.Axg7+! (7.2h4 g5!e) 7...%18 8.2xf6 x16 9.25 &xc3+ 10.bxc3 Wg5, Mest- rovic-Srebrnic, Bled 1998. Black has a bad position, but he is still able to fight on. 7.2xe7+ Wxe7 8.d5 Wes 9.2xf6 6 oxi 10.c3 All this has happened incredibly often, and now as well as on the next move it is more a matter of taste how one con- verts the advantage. But White must not be too careless, as he is still behind in development, and Black might get some dynamic chances after inaccurate play. For example, 10.Wd2!? d6 (10...Wxb2!?, as 11.Bd1 - given +— in some sources - is not at all clear after 11...d6 12.Wh6 2f5) 11.0-0-0 Se6 12.2xc7 Hac8 13.axe6 Wxed 14.b1 a5 15.Wxd6? Hfd8 16.Wg3+ Wh8 17.21 Db4 with a lot of counter- play and later 0-1, Schlenker-Nikolic, local rapid tournament 1993. 10...2e7 11.ab4 Amore technical approach is 11.\xe7+. On the other hand, White cannot hope to mate his opponent after exchanging the knights: 11...Wxe7 12.e3 dé 13.2d3 Wed 14.W13 wh8 15.0-0-0 Bg8 16.h3 S2e6 17.4b1 2d5 18.e4 2c6 19.Hhet {5 20.93 fxe4 21.2xe4 2xe4+ 22.Exe4 Wg5 23.Hde1 with an overwhelming ad- vantage, Hermlin-Auvinen, Tammisaari 1996. 11...d6 12.4\d3 Wed 13.e3 Be8 14.24 25 15.2d3 White does not and should not play over-ambitiously here. If nothing special happens, the doubled f-pawns still guarantee a win for White. Some un- spectacular manoeuvres are sufficient. 15..We5 16.2xf5 Wxf5 17.0-0 Agé 18.Ah5 Aha 19.93 Wg5 20.%b3 bé 21.2ad1 £5 22.205 372 ew mala oe a Aw ge a, a wee ey a ABS | a. Bae 22.16 23.Wc4 Hac8 24. Wd Hed 25.25 We7 26.Wh6 16 27.Axf6+ Gh8 28.Exe5 dxe5 29.01 2f8 30.axh7 1-0 The following game features two corres- pondence chess super stars: Tinu Yim from Estonia became World Champion for the second time at this 14th Wch, something nobody had achieved before. Ove Ekebjzerg was trailing by only half a point and became Vice World Champion.1.203 e5 29 He scored 10% out of 14, but 6 out of 7 in his games with the white pieces, all opened with 1./\c3! Their theoretical battle in the following game proves that the 1.03 ‘service’ can be as hard to neutralize (and that can only be done with very precise play) as any ‘main-line-1.e4’. Or, expressed differently, after the slightest inaccuracy Black irrevocably has a bad game. Game 6 Ekebjeerg - Yim corr. (14th Weh) 1994-99 1.203 e5 2.013 Ac6 3.d4 exd4 4,.0xd4 AY¥6 5.295 2b4 5...s005 is a rather solid alternative, but it offers White some scope to develop a slight initiative: 6.e3 For once, 6.45? is not good due to 6... RxXf2+!. 6...xd4 A somewhat more risky option is 6...h6 7.2h4 (72x61? Wxf6 8.e4 2b4+ 9.c3 We7, Petrik-Gazik, Slovakia 1994; 8.4)d5!? or 8.2db5!?) 7...0-0 (concern- ing 7...2\xd4, compare the comments at the next diagram position) 8.45!? (8.2e2+ Dunnington, however, now 8...Axd4 is safer as before as White no longer has Wd1-f3 at his disposal. In Skytte-A.Zude, Bundesliga 2004, Black gradually neutralized the first player's small initiative after 9.exd4 2e7 10.0-0 d5 11.203 c6 12.4e1 He8 13.Wd2 2d7 14.Be3 h7) 8...d5 9.Wi3 2xf5 10.Wxf5 &b4 11.2d3 Wd7 12.0-0-0 Wxt5 13.2xf5 &xc3 14.bxc3 De7 15.2d3, Conquest- Smejkal, Ostrava 1994. According to Dunnington, White's bishop pair is more important than the disadvantage of the doubled pawns. After 6...0-0, van Geet suggests - according to an old game of his own - 7.Qd5 (?! Keilhack) 7...2xd4 8.exd4 We8+ 9.3 Aed, | would rather judge this as equal. 7.5 and 7.%e2 come into consideration. 7.exd4 £e7 8.Wf3! In this position, which reminds one of the French Exchange or the Petroff Defence, White develops a slight initiative on the kingside: 8...c6 (or 8...d6 9.0-0-0 0-0 10. h3 d5 11.2xe7 Dxe7 12.23 d5, Suba- Giorgadze, Mondariz 1999, 13.Wh5+ Kuijf) 9.43 d5 10.h3 0-0 (or 10...2e6 11.Qe2 h6 12.23 2d6 13.493 Wad+ 14.c3 0-0-0 15.0-0 Ede8 16.b4 with the initiative, Flemming-Hirstel, corr. 2000) 11.0-0-0 He8 12.94, Bricard-Koch, France (ch) 1997. The last move also indicates why 6...h6 7.2,h4 @xd4 8.exd4 with an analogous course (S.Larsen- Bohak, corr. 2000) isn’t an improvement at all - the h6-pawn becomes a target. 6.xc6 This mainline gives White an easy game, but the winning chances are limited if Black defends accurately. Therefore, 6.3 0-0 comes into consideration: a) Inconsistent is 7.\xc6 bxc6 8.Wd4 from the game Mestrovic-Rozentalis, Biel 1990, White is a tempo down compared to the main line. b) 7.0-0-0 This involves a high strategic risk, as after 7...2XC3,30 Chapter | 8.Yxc3 Ded! 9.2xd8 Dxc3 10.bxc3 )\xd8 (T.Eckhardt-M.Uhl, Germany 2004) leads to equality at best. Therefore, 8.bxe3 h6 9.2h4 d5 (or 9...d6 10. Axc6 bxc6 11.e4 We7 12.13 d5 «/F, Dieu- Marciano, France 1998) 10.@\xc6 bxc6, Suba-Korneev, Zaragoza 1999. Suba, a remarkably undogmatic grand- master, here likes to bet on the trump ‘bishop pair’ despite such minor details as a wrecked pawn formation in front of his own king. The game continued as follows: 11.Wd4 g5 12.293 We7 1343 Hes 14.212 2f5 15.We5 We6 16.e3-, finally 0-1/4 after a see-saw struggle. c) 7.a3(!) 7.h4!? and 7.f4!? are interesting as well. 7.a3 is seemingly illogical, since in other lines Black voluntarily takes on c3, but here in particular, it prevents the ...a\e4 trick: 7....2x03+7! 8,Wxo3t Hed? 9.dxd8 ®xe3 10.2xc6 dxc6 11.2xc7 with an extra pawn (there is no d1-rook to take). 7.2.65 7.205 8.Wg3 &xc3+ 9.Wxc3 dé, Mestrovic-Yemelin, Zadar 2002 - White has the bishop pair. 8.2xc6! 8.e3 Axd4 9.exd4 2e7 10.2e2 d5=, Mallée-Kribben, corr. 2002/03. According to corr. GM Ralph Mallée, he just forgot to insert 8.2xc6 in this game. 8...bxc6 9.e3, Mallée-Prizant, corr. Petrov mem 2000-2003. White has a tiny edge due to Black’s double pawns and went on to win. The following out-of-sorts performance by lvanchuk was not a good advert for 1.2c3: 6.e37! 0-0 7.2e27! (7.1317, e.g. 7...axd4 8.exd4 He8+ 9.22 We7 10.2xf6 2xc3+ 11.bxc3 gxf6 12.02! d5 13.2d3+, Skytte-Henriques, Lisbon 2000, Black played over-ambitiously here; 7.Dxc6!?) 7...H6 8.2x16 2xc3+ 9.bxc3 Wxf6 10.0-0 d5 11.Eb1 (better is 11.c4, looking for equality) 11...8d8 12.Axc6 Wxc6 13.Wd4 2e6 14.213 b6 15.c4 Bab8 16.Eb4 We8! (threatening ...c5) 17.c5 We7! 18.c6 Hd6 19.Wc3 We8 20.04 Wxc6 with a clear extra pawn, Ivanchuk- Anand, Monte Carlo (blindfold) 1998. Finally, 6.e4?! is a minor line of the Scotch Four Knights: 1.e4 e5 2.13 Acb 3.Ac3 DE 4.04 cxd4 5.2xd4 &b4 6.2957! (6.Axc6 bxc6 7.203 is normal); 6...n6 or 6...We7 are good answers. 6...bxc6 The question arises whether Black should insert 6...2x¢3+ 7.bxc3. After 7...bxc6, White should be able to obtain a small edge, e.g. 8.Wd4 (8.e3) 8...0-0 9.Bb1 (or 9.e3; 9.2xf6 from Beikert-Marciano, France 2002, looks premature) or 8...h6 9.We5+!?. Black may defend this po- sition, but it's an unpleasant task. Therefore, 7...dxc6 8.Wxd8+ Wxd8.1.23 e5 31 White has, at least from a formal point of view, the better position. Not so much because Black has forfeited the right to castle, but because of the bishop pair as well as the sound pawn majority on the kingside. As is well-known since Nim- zowitsch’s famous book My System, a doubled pawn formation is very useful for defensive purposes, but lacks offen- sive power. Therefore: on the queenside, the three white pawns are stopping Black's four pawns. On the other hand, very recently since Kasparov's futile attempts to break down Kramnik’s ‘Berlin wall’ (1.e4 e5 2.13 DAc6 3.2b5 DFE 4.0-0 Dxe4 5.d4 Qd6 6.2xc6 dxc6 7.dxe5 AfS 8.Wxd8+ ‘Gxd8) it has become known that there are ‘better’ positions that nevertheless are impossible to win. However, there Black has the much- praised bishop pair, here White does. Back to our position. After 9.e4 h6! 10. axf6+ (practically forced, at the latest after 10.2h4 g5 the e4-pawn is left hanging) 10...gxf6 White didn’t achieve more than a draw in two games (Willke- Motwani, Gyér 1990, and van Wieringen- Simmelink, corr. 1996). Indeed, the double pawn complex f7/f6/h7 hampers the clear white majority more effectively than the sound formation f7/g7/h7 does = and knights, which could exploit the weak squares, are no longer on the board. Looked at it like that, the move 9.e4? proved to be an inconspicuous mistake; maybe the best was 9.0-0-0+ Ge7 10.13! (after 10.e4 h6 11.2h4 g5 12.f3, Rigo-Mira, Vienna 1986, 12... 42e6! is a good answer) followed by e4; the g5-bishop heads in the direction e3/f2. 7.ud4! It's very obvious that the centralized white queen has a strong position - looking towards c3 (defence), b4, a7 and up to f6 and g7. One should also pay attention to the typical 1.£3¢3 theme of fast piece development while holding back the central pawns (here e2). 7.207 The logical continuation, removing all worries about f6. If Black wanted to exchange on c3, he should have done so on the previous move (7...2xc3 8.4xc3 De4?? 9.We5+ does not work). After 7...We7, 8.xf6 is strong, 8...gxf6 9.e3 + van Geet (9...d5 10.0-0-0 2c5 11.Wh4, 1-0/30, van den Berg-Ljubarskij, Arvier 2004, 9...c5 10.Wd3 Wed 11.14 &xc3+ 12.f2! We7 13.Wxc3 with an edge, Glowatzky-Seckler, corr. 2004). 8.e4! Finally! The e-pawn does not simply make a casual move, but does so just at the moment when the further advance with e4-e8 is in the air. 8...0-0 Somewhat more cautious is 8...d6. But more cautious doesn't necessarily mean better. As the next comment proves, the advance e4-e5 isn't so much of a threat. However, Black might be playing ...d7- d5 at some point, therefore 8...0-0 is more flexible and at least preferable from a theoretical point of view.32 Chapter | The immediate 8...d5?! was surprisingly successful after 9.exd5 cxd5 10.2xf6 2x6 11.Wxd5?? &xc3+ 0-1 in a cor- Tespondence game. Nevertheless this move is dubious, e.g. 9.0-0-0 or 9.exd5 exdd 10.2b5+ &d7 11.2xf6 &xb5? (11...8xf6 12.Wxd5_ &xc3+ 13.bxc3 2xb5 14.Wxb5+ Wd7, unfortunately = only) 12.2xe7 Wxe7+ 13.¢d1!. 9.2.03 The watchword is ‘flexibility’ also for White - castling to the right or to the left side remains open. Or else: 9.e5 Dd5 10.Rxe7 (10.2d2!? Dxc3 11 Wxe3 Dunnington) 10...Wxe7 11.axd5 exd5 12.0-0-0 (12,.Wxd5? Wb4+) 12... 2b7 13.4 c5 14.Wd2 d6 15.Eet dxe5 16.Bxe5 We7 with roughly equal chan- ces in Hermlin-K.Koskela, Turku 1997. 9.0-0-0 d6, and now: a) 10.94 d7 11.2xe7 Wxe7 12.94 a5 13.h4 2c5 14.95, Pel-Wilting, Groningen 1994, with a typical struggle on different wings (14...2.g4! 15.Be1 a4 Dunnington), or b) 10.e5 gd (10...dxe5 11,Wxe5 2d6 12,Wd4 He8 13.g3 with an edge, Brodda- B.Shulman, corr. 2002) 11.2xe7 Wxe7 12.exd6 Wxd6 13.Wxd6 cxd6 14.Bxd6 x2 15.21 £ Dunnington. Besides, 9...c5 (Nunn/NCO) 10.Wd2 dé 11.147! h6! 12.2xf6 (due to 12,2h4? }xe4!) 12...2xf6F, Pell-Bellin, Monte- catini Terme 1995. To avoid those ..2xe4! tricks, preferable is 10.Wd3, e.g. 10...d6 11.e5 Agd 12.2xe7 Wxe7 13.exd6 WI6 (13...cxd6 14.Wg3t) 14 QdS Yxd6 15.We2 Wh6+ 16.14 2e6 17.93 Bab8 18.2924, Lacrosse-Meul, corr. 2002/03. 9...h6 9...d5 is playable immediately too, e.g. 10.0-0 (or else 10.e5 or 10.0-0-0!?, if White really wants a fight) 10...dxe4 14. Whd8 &xd8 12.Axe4 Axed 13,kxd8 Axi2 14.8xc7 @xd3_ %-%, Tinture- Cadillon, corr. (France ch) 2000 10.24 Or 10.:h4 d5 11.5 Dg4 12.8xe7 (12. 2g3 is well met by 12...f6) 12...Wxe7 13.f4 {6 14.0-0! fxe5 15.fxe5 Exfi+ 16. Exfl Wxe5 17.Wxe5 Axe5 18.2et Axd3 19.cxd3 Wf7 20.d4, Benjamin-Nunn, London (Lloyds Bank) 1987. Because of the weak c5-square, Black's extra pawn is of no value, finally even 1-0/97. Tinu Yim now demonstrates an accurate way to prove equality: 10...d5! 11.0-0 dxe4 More precise than 11...c5 M.Larsson-M.Lopez, corr. 2000. 12.Wxd8 Exd8 13.dxe4 @d5 The doubled c-pawn doesn’t matter, as is well-known from many 1.e4 e5 positions. But Yim even has calculated somewhat further, White's position also gets a little dented now: 14.2e5 Ab4 15.Bfd1 Dxd3 16.2xd3 Exd3 17.cxd3 2a6 18.d4 2d8 19.c5 2xc5 Finally a draw because of the bishops of opposite colours: 20.dxc5 Bd5 21..s.xc7 Exc =. YoVe 12.Wa4,1.23 e5 33 Section 6 1.Dc3 e5 2.213 Dc6 3.d4 exd4 4.2\xd4 2b4 4...2b4 is a fairly safe continuation, but it is not sufficient for full equality. Fun- damentally, the game is similar to 4...f6 5.295 2b4. Interestingly, the £g5/2f6 pin does not always help White, see for instance 6.Wd3 0-0 7.0-0-0 2xc3 8.Wxc3 Axe4!. More- over, one may compare 4...Af6 5.295 &b4 6.2xc6 bxc6 7.Wd4 £e7! (parrying everything for the moment) with 4....2b4 5.Axc6 bxc6 6.Wd4, and the attack on g7 forces either ...2xc3+ or ...2f8. This should make it clear why 4....2b4 is somewhat inferior to 4...Af6 5.2.95 2b4. Game 7 Frick — Eckert 2nd Bundesliga 1989 1.2¢3 e5 2.443 Dc6 3.d4 exd4 4.2\xd4 2b4 5.Dxc6 This technical approach is also the first choice here. However, the alternative 5.Wd3 really comes into consideration now. The goal is not acquiring the bishop pair by a2- a3, thus preventing doubled pawns (this would be too time-consuming), but 6.2.42 followed by 7.0-0-0. A sharp middlegame with mutual chances will arise, and a more definite evaluation does not seem possible for the moment: 5.Wd3 (5.93 26 6.292 0-0 7.0-0 He8=, Mestrovic-Crepan, Slovenia 1999) a) 5...d5 6.2xc6 bxc6 7.Wd4 2xc3+ 8.Wxc3 WIE 9.Wxf6 Axf6 10..2f42, but draw agreed in the game Knab-Zeltner, Germany 1991 - it's an important conclusion that ...d5 ‘with a grip on the centre’ doesn't work so easily. b) 5...WH6 6.2db5 2a5 (6...2xc3+? 7. bxc3 Wd 8.Wg3, Hainzinger-Gutdeutsch, Austria 1995, 7...We5? 8.14) 7.Wg3 (7.242 Qge7 8.0-0-0 a6? - 8...0-0 - 9.De4 2xd2+ 10.Wxd2 Wed 11.Ded6+! cxd6 12.4\c7+ Yd8 13.2xa8 b5 14.2b6 &b7 15.e4+-, Mestrovic-Zecevic, Be- lisce 1999) 7...d6 8.2d2 Age7 (8...a6? 9.Dxc7+ 2xc7 10.0d5 Wd8 11.Yxg7+-) 9.0-0-0 a6? (9...2b6) 10.e4 We6 (10. Wg6 11.2xa5) 11. Wxg7 Eg8 12.0xc7+ 1-0, Aasum-Ekmark, corr. 1985. c) 5.246 6.295 (6.Axc6!?) 6...d5 (6...0-0 7.0-0-O%) 7.@\xc6 bxc6 8.a3 e7 9.63 0-0 10.2e2 2e6 11.0-0 Wd7 12.Wa6 Hfb8 with mutual chances, Unger-Kempter, Germany 1996. d) 5...2ge7 (which | feel to be the best move) 6..2d2 0-0 7.0-0-0 (7.64 d6 8.2\xc6 @xc6 9.0-0-0 a5, Grabenhorst- Grajetzky, corr. 1990) 7..d5 8.A\xc6 bxc6 9.e4 £e6 with mutual chances, Krajnak-Tepper, Slovakia 1997. 5... 2KC3+ On 5...0xc6 follows - as already mentioned - the thematic move 6.Wd4!. 6....2f8 in- volves a suspicious-looking delay in de- velopment, therefore Black has to take on c3 without creating doubled pawns. After 6...2xc3+ 7.Wxe3 Wf6, the dry 8.Wxt6 is already sufficient to prove an advantage (+ bishop pair, pawn34 Chapter | structure). More substantial is 8.Wa5!? Wg6 9.2d2 Se7 10.e3 0-0 11.2d3 Wxg2 12.0-0-0 with good attacking pro- spects, van Geet-Apking, Amsterdam 1960, or 8.2d2! Wg6 9.93 Se7 10. &g2 0-0 11.h4 d5 12.h5 Wd6 13..2f4. baa 240 Wal ef In the correspondence game P.Hoffmann- Okruccinski, 1994, Black resigned at this point. Indeed 14.h6 is very strong, the more so as Black additionally is aff- licted with his pawn weaknesses c6/c7. 5...dxc6 6.Wxd8+ dxd8 7.2d2 266 8. 0-0-0 &c8 may well be playable for Black, but should be rated =. Unlike in the Ruy Lopez (Exchange or Bertin Defence), White can lay claim to his kingside majority without having to cede the bishop pair. 6.bxc3 bxc6 After 6...dxc6 C.Frick, in his game against Kunert, Germany 1999, applied his pet move 7.Wd4. After 7... Wxd4 (7...2)f6) 8. cxd4 &e6 9.e4 Black obtained counter- play against White’s centre: 9...5 10.f3 0-0-0 11.2e3 Af6 12.203 fxe4 13. fxed Bhe8 14.2, and now 14...c5!? 15.c3 (15.d5? 04! 16.dxe6 Axed+ o) 15...cxd4 16.cxd4 4\g4. was possible (G.Schnepp in the tournament bulletin) - this reminds us of the Griinfeld Defence. Still, 7.8xd8+ is simpler; | refer to the discussion of a similar position on the occasion of Game 6, 6...2.xc3+. 7.4 The thematic move! Actually, the question is whether this or 7.g3 W16 8.Wd4 (as was actually played in the game) is the more precise move order. After 7.3, 7...2f6 is also possible, e.g. 8.243 d6 9.292 2d7 10.0-0 0-0 11.Wd3 with a tiny edge due to the bishop pair, van Bellen-P.Kramer, Netherlands 1966. Anyhow, Black's position is tenable. In fact, 7.Wd4 is partly a bluff: 7...f6!?, and now neither 8.295 (resulting in an endgame with doubled f-pawns, however, the bishop pair balances this. At least, the b-file is a small trump for White: 8...0-0 9.e3 d6 10.2b1 h6 11.2xf6 Wxfé WKIE gxf6 13.ed2 Se6 14.4b72, Stancl-Portych, corr. 1999) nor 8.We5+ (after 8...\We7 9.Wxc7 0-0 Black obtains compensation; ideas such as ...WWc5, ..2d5, ...De4) constitute a real threat. Generally, White is hoping that the second player, facing such unpleasant moves as sg5 or We5+, loses sight of the real long-term danger - namely, losing an endgame against White's bishop pair! Occasionally, a pseudo Scotch with 7.e4 is played - not a really convincing idea. The approach with 7./8.Wd4 is much more in the spirit of the position. 7... WH 7...216!?, as mentioned.1.2c3 e5 35 8.93 8.24 d6 9.2b1 25 with the remarkable sequence 10.%&d2! éd7! 11.8 2e6 12.a4 C5 13.Wxf6 Dxf6 14.e4 Hab8 15.203 in Schéllmann-H.Jérg, Bavaria 1993. In view of the idea of e4-e5 as well as White's bishop pair, Black has not yet equalized. 8.h4!?, Stinis-Hass, Leuven 2003, 8.6b1 d6 9.93 Wxd4?! 10.cxd4 Ae7 11.c4 2a6 12.e4 0-0 (12...d5!?) 13.243 Efe8 14. d2 +, ‘RedBishop'-Boniek, corr. 2002. 8...d5 Looks clumsy. However, a setup with ..d6 does not relieve Black of his wor- ries about the double pawn formation, as the following examples show: 8...Wxd4 9.cxd4 246 (or 9...d6 10.292 @e7 11.8b1 0-0 12.0-0 2a6 13.He1 Hab8 14.4d2 2c4 15.2a5t &xa2? 16.Exb8 Exb8 17.2xc7 Eb2 18.2xd6 cB 19.2c5 Exc2 20.2xc6, van Bellen- Belinfante, The Hague 1975, with a winning endgame - the strength of the bishop pair!) 10.2f4 d6 11.292 De7 12.0-0 0-0 (12..d7!? 13.2h3+) 13.2d2 Hab8 14.ofb1 Dd5 15.e3 £5 16.82a5 &f7 17.b3 2c4 (by means of an exchange sacrifice, White avoids bishops of different colours) 18.2xd5+ &xd5 19.2xc7! 2xb3 20.axb3 Eb7 21. 2xd6 Ha8 22.205 a5 23.Ha4, Mestrovic- Srebrnic, Ljubljana 1993. With two pawns for the exchange, White has the upper hand, as Black's rooks have no prospects (1-0/90). From a formal point of view, White's iso- lated double pawns c2/c3 seem weaker than Black's triplets on the c- and d-file. On the other hand, a weakness which cannot be exploited by the opponent is not really relevant. Here, White’s splen- didly centralised queen - and after it is exchanged, the king! - easily cover this weakness. 8...b8 9.2.92 Hb6 10.0-0 d6 11.2634, Glowatzky-Bolaf, corr. 2004, doesn’t solve Black's problems (a7, c6, 97), e.g. 11...2a6 12.8 fb1. Interesting is 8...%g6!? (eyeing up c2) 9.292 De7 10.e4 (10.Wed 16 11.Wxc7 Wrxe2) 10...d6 11.Hb1 0-0 12.0-0 Web 13.Wa4 a5, and Black held his own, Roques Pascal-Cereda, corr. 1999. 9.214 With the clumsy but effective threat 10.2e5. 9...Wxd4 Also after 9...Wg6 10.c4 “f6 11.cxd5 (+) Black's equalising attempts are futile, no matter how he recaptures on d5. 11.02! As indicated above! 11...2b8 12.3 From here, the king covers the entry square b2. 12...Af6 13.43 ted7 14.292 Hb6 15.hab1 EZhb8 16.2b3 h6 17.2e1 | would have peferred 17.94!?, with play ‘on the kingside. 17...2.06 18.2.1 But this re-grouping is not bad either. 18...a5 19.2.a3 De8 20.2.5 a4 21.Exb6 cxb6 22.5b1 Se7 23.21836 Chapter | The bishop dominates the e8-knight - until it is finally exchanged for the said piece. 23...n5 24.b4 Ha8 25.a3 Ha5 26.4b1 g6 27.03 b5 28.2203 Dd6 29..2.xd6+ Transformation of advantages - now it's ‘good’ against ‘bad’ bishop. However, in spite of the indisputable edge which White has, Black needs to commit some errors to make it a clearly winning position. This is a typical occurrence in such endings, which in practice are mostly lost. 29...exd6 30.e4 6 31.2e1 a8 32.f4 dxe4 33.2xe4 (5? 33... 98 or 33...,f7 was necessary. 34.213 2d7 35.b4 ha? The common nervousness in time-trouble, which often leads to strategic errors. The second player wants to get the h-file..., 36.gxh4 Eh8 37.h5! ... however, after this tit-for-tat answer, the first player gets the g-file. 37...gxh5 38.2g1 Eh6 39.h4 28? Too passive (39...2e6!). 40.95 296 41.c3! Due to his previous mistakes, Black now is in zugzwang! (41...%c7 42.25) 1-0 Section 7 1.23 e5 2.23 Ac6 3.d4 exd4 4.2\xd4 2c5 After 4....205, White has various tempt- ing continuations; it’s hard to say which one is the strongest from the theoretical point of view. White obtains a very slight advantage in all three cases, therefore his taste must decide between 5.2\xc6 (and 5.2b3) > Game 8, 5.f5 > Game 9and 5.2e3 > Game 10. Game 8 Ekebjzerg - Palmo corr. 1989 1.203 e5 2.13 Ac6 3.d4 exd4 4.2xd4 25 5.2\xc6 The resulting position with the doubled pawns has a somewhat different character compared to the Scotch Game, where White is so to speak ‘compromised’ by e2-e4. (it goes without saying that this is not just a ‘disadvantage’ in the Scotch - instead White builds his plans around the move e5 and/or a supplementary f4) In our line, a possible ...d7-d5 will be met by c2-c4. The game reminds us more of the English or the Catalan Opening or a Grinfeld with colours reversed. Another possibility is 5.2b3. This move is not bad either, but it offers, comparatively speaking, the smallest chances for an opening advantage 5.5.6 5...2b4 6,Wd3 (6..2d2 D6 7.a3 2e7 8. &f4 d6 9.e4 26 10..2b5 =, Kritz-Siklosi, Paks 1999) 6..ge7 7.43 &xc3+ 8.Wxc3 0-0 9.295 h6 10.ah4 Wes, Bellon-Westerinen, Dortmund 1980.1.23 e5 37 White has the very slight advantage of the bishop pair. The following sequence was absolutely exquisite: 11.94!? d6 12. h3 a5 13.292 a4 14.d2 2d7 15.2161? gxf6 16.de4 Ads 17.Wd2 Wed 18.Wxd5 Qd4 19.5d1 Axc2+ 20.811 Yxd5 21 Qxt6+ Sg7 22.axd5 HaS 23.2f3 Bb5 24.Dxc7 Exb2 25.8xd6 &c8 and finally Ya-Val53. 5...2e7 6.93 Df6 7.292 is illogical, compare 4...af6 5.93 &c5 6.ab3 2e7 (Game 5) 6.205 Consistent. 6.93 and 6.24 are playable, but do not promise any opening ad- vantage. 6.e4 > 1.64 e5 2.413 Ac6 3.d4 exd4 4.Oxd4 205 5.2b3 2b6 6.c3. 6... 46 7.2\xb6 (7.2.95? 2xi2+) 7...axb6 8.93 Or 8.295 h6 9.ah4 Had 10.893 0-0 11.e3 d6 12.c3 We7 13.2b5 Ha8=, As.Hoffmann-Finegold, New York 1985. 8...d5 9.2g2 0-0 10.0-0 2e6 11.404 Qxd4 12.Wxd4 Wd7 13.ce1 5 14.d1 2h3 15.ah1 Bfes, Ban-Kluger, Budapest 1956. Based on the superior control of the centre, Black has his fair share of the play. However, the white trumps - the bishop pair and a healthy pawn structure - also count for something: at least =. 5...bxc6 The jury is still out on 5...Wh4 6.93 Wi6, probably 7.£14 dxc6 8.£g2 is quite favourable for White. 5... WHE 6.2e4 Wxc6 7.2xc5+, bishop pair. 6.93 6.e4 leads to a variation of the Scotch Game (1.e4 e5 2.43 Dc6 3.d4 exd4 4.Axd4 £05 5.2xc6 bxc6 6.203), where 6.44c3 (6.2031) is not the first choice - the reply 6...Wf6!? shows the reason why. 6...Df6 7.2.92 0-0 8.0-0 8...a5 Black has several alternatives, but they won't change the character of the position very much, e.g. 8...2b8, 8...2a6 9.044 2d6 10.b3 He8 11.c4 (K.Wilhelm- Wochnik, corr. 1998), or 8...d5 9.2a4 2d6 10.2e3 He8 11.c44, van Geet- Kunnen, Rotterdam 1960. The exchange sacrifice 11...Bxe3 12.fxe3 Dg4 13.23 might have been playable as a long- term positional sacrifice (13...2e6 or 13...dxce4 14. Wd4 &e6 with some compensation), but after the hectic 13...2xh2?! there followed 14.%xh2 Wh4+ 15.891 2xg3 16.Exg3 Wxg3 17.We1 Wxe3+ 18.Wt2+ etc. 9.204 £a7 10.c4 In the English Opening, there is something similar with 1.c4 e5 2.c3 BNC 3.213 g6 4.04 exd4 5. Dxd4 seg738 Chapter | 6.Axc6 bxc6 7.93 De7/Af6. The little hardship, that White first has to remove the c3-knight, unblocking the c-pawn, is more than outweighed by the fact that Black's dark-squared bishop is by far not as effective on a7, as it would be on g7. 10...d5~ With the c7-pawn on the e-file, one could look upon this as a Grunfeld Defence with colours reversed. In spite of the semi-open e-file, this comparison also turns out rather well for White. 11.295 dxe4 On the other hand, the tripled c-pawns are characteristic of the Catalan Opening, for which one should add &d4 and Ae6. The corresponding lines are 1.d4 d5 2.c4 e6 3.13 AWE 4.93 dxc4 5.292 £e7 6.0-0 0-0 7.He5 6 8.Axc6/2xc6 bxc6 or 7.2 a6 8.a4 2d7 9.8d1 &c6 10.203 &xf3 11.2xf3 Deb 12.2xc6 bxc6; the piece play compensates Black for the weak pawns. A point in favour of ‘our’ version is that White already has a 4:3 pawn majority at the kingside, whereas in the Catalan lines, there isn’t any visible majority, 12.Wc2 2e6 13.2xc6 Hb8 14.2ad1 We7 15.We3 2h3 | will often express myself on how em- barrassing it is to annotate Ekebjzerg games! Black's whole play in this game seems rather logical, and apart from the final blunder | do not have a clue what he could have done better! 16.2xf6 Wxf6 17.Wxf6 gxf6 18.2fe1 2e6 19.2d2 Hfds 20.Hed1 Hxd2 21.Exd2 &f8 22.492 we7 23.e3 2f5 24.D¢3+ 2.052? 25.8d5 1-0 Game 9 Timman — Hibner Candidates match Sarajevo 1991 1.2¢3 e5 2.213 Dc6 3.d4 exd4 4.0xd4 205 5.25 The Scotch - Schlenker's discovery, of which the 1.23 community is very proud. In the Scotch, after 1.e4 e5 2.13 2c6 3.d4 exd4 4.2xd4 2.c5, for some time 5.215 was fashionable. White had favourable experiences after 5...Wf6 6.2e3 (£ according to theory); however, the line soon disappeared after it was found that 5...d5! promises strong counter-play (by the way, 5...d5 was by no means new but a well-forgotten sug- gestion by Steinitz). However, more inter- esting for us is the statement that 5...96 6.4e3 is also playable, the assessment varies between ‘’ (NCO) and ‘=" (ECO).1.23 e5 39 Now, the hyper-compatible train of thought runs as follows: (1.Qc3 e5 2.218 Dc6 3.d4 exd4 4.4xd4 205) 5.5 practically ‘forces’ the response 5...Hf6, as the counter-stroke 5...d5 is ineffective without a pawn on e4. After 5...Wf6, White simply plays 6.e4, entering a favourable line of the Scotch - which, due to 5...d5, cannot be reached any more in the true Scotch, but can via 1.4c3!. There is only one problem - 5...¥8/16 is not really as forced as some authors claim. 5... Wi6 In fact, this has been played in the vast majority of games. There is not much material concerning the other moves, therefore, to draw some provisional conclusions, we must compare the resulting positions with the original Scotch: a) 5...d5? 6.2xg7+ &f8 7.Qh5 Wh4 8.293+ was gladly played in some games, wrongly in analogy to the Scotch: 1.e4 e5 2.4f3 @c6 3.4 exd4 4.2xd4 205 5.4f5 d5! 6.2xg7+ ef8 7.4n5 Wh4 8.493 D6. There, Black has a very dynamic position and the missing pawn carries no weight. Here, d5 only causes headaches. Therefore, somewhat better is b) 5...d6 6.2axg7+ Sf8 7.Qh5 Wh4 8.e3 294 9.2e2 Ahé6 (Petri-Behrens, GieBen 1996) or 7...294 8.2193 h5 (Zhi Chang-Berendonk, Kleve 1999). In my opinion, Black has some kind of ‘dirty’ compensation - promising in lightning or rapid play, but not fully sufficient from an objective point of view. c) 5....2f8 6.e4 g6, and now White has the choice between two possible setups: 7.Qd4 297 8.2e3 or 7.e3 297 8.Aed5 (diagram). The first one seems more natural, the second one is more characteristic for 1.2\¢3. Indeed, some authors of books on the middlegame warn about such “doubled knights”, but here Black cannot so easily take advantage of this, e.g. Qd4? (A ..C6) 9.295+-. In Willke-Rippinger, Ger- many 1991, White obtained very strong play after 8...Age7 9.2905 0-0 10.4 W@h8 11.h4. But much better is 8...2\ce7! (the g8-knight protects some key squares, and White must reckon with ...c7-c6 at any time) with chances for equality. d) After 5...g6, the 1.4c3 literature makes us believe that this is ‘+’, bringing up games like 6.2\g3 244 7.Ad5 Age7? 8.Wxd4 1-0, Augustin-G.Haag, corr. 1987, or 6.416 7.e4 dé 8.295 Ded 9.f4 Aeg4 10.Ad5 212+ 11.e2 2xg3 12.hxg3 c6 13.Axf6+ Dxf6 14.Wd4 1-0, Huzen-A.Wagner, corr. 1997. To put things in order: d1) 6.2g3 26 7.e4 is more or less the same as 1.e4 e5 2.43 Ac6 3.d4 exd4 4.Axd4 205 5.515 96 6.Ag3 (6...A16 7.4c3), but everybody plays 6.e3 in the Scotch line, which is still not over- whelming strong (+ to =, as mentioned). Indeed, the g6-pawn nicely restricts the g3-knight - in our main game, the same method will be applied by Timman (10.g3!), and in the Van Geet Attack, the move ...g6 is a suitable measure as well. To be more concrete, in the above- mentioned game Huzen-Wagner the move 8.295?! would have been a let-40 Chapter | down after 8...h6!. The superior 8.2e2 may be answered by 8...2)d4!?. A sensible try seems to be 7.a\ge4 @xe4 8.2xe4 £04 9.03 2g7 10.295 16 11..2f4, Le Bled-Le Clanche, corr. 1997, but this is nothing special of course (rather equal). d2) 6.e3, and in comparison to the Scotch (1.e4 e5 2.23 Ac6 3.d4 exd4 4.Axd4 205 5.45 g6 6.\e3, mostly continuing 6...Af6 7.Ac3 0-0 8.2d3 Be8), the blocked e-pawn is not a piece of jewellery. White is too limited in his choice of possible plans: soon he has to play Sed5 and/or g3, instead of the natural flow of the game in the Scotch line with Ae4+ Se. In fact, after 6...016 7.93 DeS 8.ed5 Deg4 9.e3 c6 10. Axf6+ @xf6 11.892 d5 12.0-0 0-0 13.b3 Ee8 14.2b2 2f5 15.h3 h5 the game tended to a ‘F’ in Kovarcik-Flear, Saint-Affrique 1999. This is a typical scenario: a grandmaster, such as GM Flear in our case, gets con- fronted with an ‘unorthodox’ opening; of course he doesn't know the ‘theory’ and therefore chooses a line which is consider- ed inferior. He makes some natural moves and, in spite of the theoretical ‘verdict’, he soon gets a promising position. Therefore, professional players don't take the bulk of books about gambits and rare openings really seriously. On the other hand, the experts in the unorthodox prefer to stay in their own misunderstood world, and the mutual prejudices continue. The real shortcoming of the unorthodox openings seems to me a different one: The problem of the so-called “Unorthodox Openings” is not their inferiority but their lack of thorough and reliable theory. Often, the advocates of unorthodox lines are a bit too enthusiastic, while at the same time lacking general knowledge of standard positions - which often leads to horrible misjudgements in the relevant literature. On the other hand, it’s a well- known fact that mainstream theory (Encyclopedia, etc) is full of superficial pseudo-refutations of ‘minor’ lines and therefore not really helpful either. d3) Most appropriate to the situation (blocked e-pawn, providing sound de- velopment for both bishops) seems to me to be: 6.2e3! &xe3 6...&2b6!? 7.2xb6 axb6 7.2\d4, 6...218 7.d4 &Q7 B.AdS Age7? 9.895 sxd4 10.Yxd4! 0-0 11.2f6+ Bh8 12.4944 1-0, Hermlin-Kaunonen, Karhula 2000 - in- structive knight tactics by the underesti- mated Estonian 1.4¢3 expert Hermiin. 6...2b4 7.Ad4 Age7 8.26 ds 9.2xc6 bxc6 10.Wd4 Hg8 11.2d2 &xc3 12.bxc3 Wi6 13.e4 with an edge, Keilhack-N. N., from an internet blitz game 2002. 7.2xe3 with the plausible continuation 7...16 8.93 0-0 9.2.92 d6 10.0-0 2e6. Now, that’s not at all a Scotch Game but a typical 1.2c3 according to the inter- pretation of Ekebjzerg: although White has no direct threats and no obvious advantage, he has more control over the game. OK, others will call this rather dull, there- fore a more interesting alternative is: 8.94!? d6 9.h3 0-0 10. Wd2 He8 11.292 b8 12.0-0-0 2d7 (I don't like this move)1.43 e5 41 13.95 Dh5 14.Ded5 Aes, Weise-Bajra- movic, Germany 2003, 15.4e4+. One might improve on this game, but 8.94 definitely breathes life into this line Back to 5...Wf6: 6.e4 As you might have already guessed, Timman and Hibner did not really play a 1.43 game but 1.e4 e5 2.13 “\c6 3.44 exd4 4.0xd4 205 5.215 WI6 6.203. Interesting and played in a very modern fashion was 6.94!? 2b4 7.2d2 2ge7 8. e4 Axf5 9.gxf5 Wh4 10.We2 dé 11.25 (11.0-0-0 Dunnington) 11...2xd2+ 12.xd2 &d8, Brduning-Bibik, Germany 1993. 6...age7 6...96? 7.25 Wed 8.2f4+-, Glowatzky- Howey, corr. 2003/04 7.2De3 Black's trumps are his actively posted pieces. However, White is threatening to throw back the black pieces in the medium term. If he succeeds in this, he will, assisted by his space advantage, control the whole game. Often a direct attack on the king results from this, e.g. by means of f4-f5-f6. Black has to play as aggressively as pos- sible to keep his pieces in their ‘floating’ posts, or to force some concessions from White (e.g., some weakening pawn move which might be the starting point for another action). A quiet approach with a ‘there-is-nothing-going- on’ attitude (playing ..d6 and ...0-0, as happens in the majority of amateur games!) seems inappropriate to me. After Black has lost the momentum, he will inevitably be steamrollered by White. Two relevant examples after 7...0-0 (or 7...d6?!) 8.2.d3 d6(?) 9.0-0: a) 9...2e6 10.8h1 Se5 11.22 A5c6 12.f4 Wd4 13.203 Db4 14.b5 WI6 15. @xc7 Hac8 16.2\xe6 Wxe6?! 17.15 Wd7 18.16 gxf6 19.294 Web 20.2\xf6+ whs 21.Wh5 1-0, Hagemann-McClelland, corr. 1992/93. b) Ag6 (better 9...e5 10.%h1 6 11.4 @xd3 12.cxd34, Schlenker) 10. DedS Wd 11.65 2xf57! (11...Dge5 12,Wh5 2xf5 from den Herder-Kersse- makers, corr. 1990, is the lesser evil) 12.exf5 ge5 13.16 (+- Schlenker) 13... QXf6 14.85 Dd4 15.Whd g6 16.Het (threatening Qe7+) 16...2xf5 17.Wxtd bg7 18.Be7! 204 (18...Axe7 19.xt6+ g8 20..2h6 DFS 21.Wxf5+~ Schlenker) 19.Exc7 He8 20.2g5! (among other things, Ed7 is threatening) 1-0, Schlenker-Hillesheim, Tubingen 1979. 7...0-0 8.2.43 8.93 eS (or 8...d5!? 9.exd5 De5 10. f4 Bxe3 11.2xe3 Dg4 12.804 Wh with a complicated struggle, Huzen-Valverde, corr. 1997) 9.f4 &)5g6 10..2g2 c6 (10... Wh6 11.Dcd5 DxdS 12.axd5 Wd6 13. e5 He8 14.8f1 Wa6+ 15.Wd3 Yxd3+ 16.cxd3 2.b6 17.2xb6 axb6 18.%f2 [6 19.44 Had etc, Ghinda-Przewoznik, Naleczov 1979. The motto: never let the first player do his thing in peace! - the continued bothering of the opponent is not by chance but the appropriate method) 11.0-0 d5 12.2h1 (12.65? xe5 13.fxe5 Wrxe5coA...d4) 12...2xe3 13.2xe3 dxe4 14.Wel! (xb2, 14.2xe4 £h3) 14...8f5 15.Dxe4 axe4 16..axe4+/bishop pair, Ljubojevic-Karpov, Montreal 1979.42 Chapter | Yet the possible idea ...d7-d5 shows why the stereotypical move ...d6 is too superficial. 8...De5 The fact that in spite of the ‘correct’ active approach things are rapidly going downhill implies that this line is de- finitively bad for Black. That leaves the small inconsistency with 5...g6, other- wise one could call 5.2f5 really strong. 9.22! Saving the ‘bad’ bishop from being ex- changed - especially since bad bishops often protect important squares. 9...5g6 After 9...d6, progress in the spirit of the above mentioned games Hagemann-Mc Clelland, Schlenker-Hillesheim is highly probable 10.g3! d6 11.h4! Outstanding preventive play against pos- sible black piece play on the kingside! Steinitz in his day already suggested the technique of restricting the opponent's knights by pawn moves. By the way, the last move also restricts the c8-bishop (no ....2.h3). 11...e8 12.h5 f8 13.2h4! Now even the king's rook moves out in order to harass the black queen! Obviously, Black is on the retreat. 13...c6 Preferable was 13...a6, Timman. 14.2304 Wd 15.2\xc5 Wxc5? After this, Black's queen is indeed lost. It was necessary to play 15...dxe5 16.2¢4+ Timman. 16.2¢4 Hd8 16...d5 17.23 Wb5 18.a4 Wa6 19.2d6 Wa5+ 20.2d2 Timman. 17.203 Wb5 18.24 Wa6 19.64 d5 20.4b6 Wxb6 21.2xb6 axb6 22.exd5 There follows a prolonged technical part: 22...2'5 23.4 Exd5 24.2.d3 g6 25.hxg6 hxg6 26.4e4 £e6 27.Wd2 Dd4a 28.We3 c5 29.bxc5 bxc5 30.24 Eh5 31.2xe6 Dfxe6 32.5h4 + 1-0/41. Game 10 Charousek — “A. S.” Vienna 1897 1.2¢3 e5 2.13 Dc6 3.44 exd4 4.2xd4 2.05 §.2.e3!? An interesting alternative, which again emphasizes the technical features of the position. Because of the threat towards the c5-bishop, Black is forced to react.ADc3 e5 43 5... 2x04?! Alternatively: 5...27?! 6.2d5! (threatening 7.4405) will at the very least result in +/bishop pair. 41 is illogical, the additional tempo ‘&c1-e3 in comparison to the 4...2¢5 line is quite useful. After 6.xc6 bxc6 7.ud4 2xc3 8.¥xc3, White has a con- siderable advantage, as is underlined by 8...W16?? 9..2d4 or 8...f6 9..£xa7!? (both moves have been played already!). Some- what better is 6...2xc3+ 7.bxc3 bxc6 8. Wd4 WE 9.93, a bold ‘t’ in the game van der Spek-Gabriels, Wijk aan Zee 1989 (White has the additional tempo £3 in comparison to Game 7). The only correct answer is 5....2b6!. After 6.2xc6 bxc6 7.2xb6 axb6 8.e4 (asses- sed ‘2’ in some sources) 8...Wg5!?, von Alvensleben-Kanonenberg, 1989, White hardly has an advantage (9.h4!?). The alternative is 6.2d5, e.g. 6...Af6 (6... Qge7 looks more logical, 7.\xb6 axb6 8.93 0-0 9.292, T.Lang-Wehmeier, corr. 1995, or 8.295 0-0 9.e3 Dixd4 10.Wxd4 d5 11.203 f6 12.2f4 c5, D.Fischer-Cas- tillo, corr. 2003; Black has some central superiority to balance out White’s bishop pair) 7.2xb6 axb6 8.2xc6 bxc6 9.295 0-0 10.Wd4 dé 11.2xf6 Wxf6 12.Wxf6 gxf6, Matzing-Bacholke, Werther 1999. The resulting endgame is hard to assess. Black's pieces are more active, his pawns are somewhat more centralized. But maybe in the long run, the clean structure of the white pawns is the more important argument. 6.2xd4 Dxd4 7.Uxd4 t+ WH 7...246 8.We5+ We7? (8.218) 9.Wxc7; however, even more convincing than taking away Black's castling rights may be 8.e4 d6 9.0-0-0 0-0 10.e5, Mestrovic- Truta, Ljubljana 1994. 8.We3+! 8...d8 Practically forced: 8...We6/We7 9.Wg3 with a double attack against c7/g7; 8... De7 9.2d5 Wd6 10.0-0-0 is also more than suspicious looking. 9.0-0-0 A\e7 + 10.g4!? Avery modern looking move! 10...2e8 11.292 c6 12.e4 Dds Otherwise Ad6 will follow. 13.2xd5! cxd5 = 14.\xf6 Exe3 15.fxe3 gxf6 Charousek finally demonstrates convinc- ing endgame technique: 16.811 we7 17.2xd5 d6 18. g5 f5 19.e4! f4 20.Exf4 2e6 21.8h4 S98 22.Eh5 ig7 23.h4 294 24.0h6 &xe2 25.2xb7 204 26.205 1-044 Chapter Il Chapter Il — The Van Geet Attack 1.2¢3 d5 2.e4 d4 3.2\ce2 The Van Geet Attack 1.4\c3 d5 2.e4 d4 3.2)ce2 followed by “\e2-g3 is the core of the 1.2\¢3 opening. While elsewhere there are constantly possible transitions to the patterns of standard openings, the Van Geet Attack is a fully independent entity which strives for early knight activi- ty on the kingside. White usually plays @g3, @f3, 204 or &b5, d3, 0-0; while Black can vary the formation of his pawns and minor pieces in many ways, occasionally he even cas- tles queenside. Due to the closed struc- ture of the position, move-orders can vary. One should know the concrete lines after 3...e5 4.g3 &e6, and besides, one should be aware of the problem of finding the right timing for f2-f4 in the 3...c5 line. For all other cases, it's much more impor- tant to develop a good understanding of the strategic issues. E.g., Section 10 is per se non-theoretical, as both parties may vary their moves in nearly any way (but it's nearly always good for White!). Therefore, | have included a lengthy in- troduction to recurring strategic issues. This seemed necessary solely in the current chapter, as otherwise the mutual strategy is mostly obvious, but this one has a bold distinguished character, including various far from obvious manoeuvres. Possibly, this is caused by the fact that the Van Geet Attack has all characteristics of a closed opening, while all other lines could belong to the semi-open games. 2...d4 is chosen either by somewhat naive players who are attracted by the fact that Black wins time and space (usu- ally going for the clumsy c5/e5/.d6/16 setup afterwards), or by strong players who are aware of the strategic risks but are striving for a complex battle. Overview: Section 8: Strategic elements of the Van Geet Attack Section 9: Early deviations Section 10: 1.23 d5 2.e4 d4 3.2\ce2 e5 4.Ng3 various Section 11: 1.23 d5 2.e4 d4 3.Ace2 e5 4.93 £e6 Section 12: The 3...c5 system - 1.2\¢3 d5 2.e4 d4 3.2\ce2 c5 Section 13: The Lizard Attack - 1.2c3 d5 2.e4 d4 3.\ce2 followed by 92-93 Section 11 with 4...2e6 is the theoretical mainline, while Section 13 shows a totally different approach, far from the intentions of the Van Geet Attack.The Van Geet Attack 1.03 d5 2.e4 d4 3.2\ce2 45 Section 8 Strategic elements of the Van Geet Attack We'll look at the following elements: Basic (knight) tactics Sudden attack on the king The thematic lever f2-f4 Exchanging bishops - some thoughts about ‘good’ and ‘bad’ bishops The march of the h-pawn: ...h5(-h4-h3), The black pawn wall on the kingside The doubled c-pawns The phantom pawn on d3 The pawn push b2-b4 ‘Defensive energy’ and sliding into a messy position The discussion is not finished with this section. Some of those elements are the focus of attention in some critical lines. 698869 SeE80 Besides the above, the motif c2-c3 should be mentioned. White usually resorts to this measure when Black has prepared himself well against other plans like f2- f4. Compare the King’s Indian, where (after d4-d5) ...17-f5 is Black's standard thrust, but occasionally he plays ...c6 as well. The logic behind this is hard to grasp, as in the King's Indian, it's usually White's task to take action on the queenside. It's hard to illustrate c2-c3 clearly, as it's often only an adjunct to other plans, and its influence on the game is more subtle Concretely, c2-c3 indirectly helps to sta- bilize the position of the c4-bishop (in the King’s Indian, Black often plays ...a5 and ..c6 when he has a knight on c5), in some cases, it makes b2-b4 (item ©) work; and it’s particularly beneficial to play c2-03 when Black has blocked his c-pawn by playing ...2c6. In this case, White has a good chance to gain superiority in the centre. @® Basic (knight) tactics The following may look rather suited for a beginner's book. However, after the unaccustomed 1.4\c3! such disasters are not at all rare: 1.203 d5 2.e4 d4 3.2\ce2 e5 4.293 c5 4...06 5.23 2d6 6.2.c4 Age7??. This position was reached, via a ‘small’ trans- position (1.e4 e5 2.43 d5 3,031? d4 etc), in the top-flight Internet Chess Club encounter Viassov-Radjabov, 8th Febru- ary 2003 (the latter was just warming up for Linares where he was going to beat Kasparov). Unsuspecting, White contin- ued with 7.0-0, missing the chance to crush the youngster with 7.295! 0-0 (7...8f8 8.xh7) 8.Wh5 +-. 5.2.04 Or 5.413 294 6.204 16? 7.2\xe5! &xd1? 8217+ we7 9.515#, Moody-S.Bender, corr, 1985. 5.16 5...2e7? 6.Wh5 g6 7.Wxe5 +, Leeners- Douwes, match 1972. 6.093 £942? T.2x87+ This motif finds its victims again and again. We? 8.2xe5! And now even Legal’s mate: 8...2xd1 9.515#. The move 7.A\xe5 was already possible in the diagram position (7....£2e6). 8...Wc7 9.2xg4 and 1-0/41, van Bellen- Glende, Wildbad (Wch veterans) 1993.46 Chapter II The following game shows knight tactics in a less trivial manner. The victim, the late Hungarian IM Cserna, was a well- known open specialist at the time: Game 11 Frick —- Cserna Vienna 1984 1.2¢3 d5 2.e4 d4 3.ce2 25 4.3 As we will see in Game 17, this is not the best move. Also the next moves of both parties are far from perfect. But in the year 1984, even the few specialists could not know everything. 4...206 5.0g3 2e6 5... Af6 6.2.04 2.94 7.d3 Dn5? 8.2x17+! Sxf7 9.Axh5 &xh5 10.Axe5+ &xe5 11.WxhS+ web 12.Wi5+ Sd6 13. 4 +— (1-0/19), Schleef-Menke, 1988/89. 6.2.b5+ Preferable is 6.c3 first. 6...2d7 More logical is 6...c6 7.244 Sa6 or 7... Qd7 (7..%¥a5 8.c3!). But the situation after this is less favourable for Black than in Game 22 - here, the d6-bishop obstructs the d-file, which is a disadvan- tage in view of the typical motif ...d4-d3!. 7.2.04 7.€31 ¢5 8.4 would transpose to Game 24. 7...c5 2! according to Moser, but | think the move isn’t that bad. 8.0-0 16 9.d3 De7 10.2b3 @18? But this seems too artificial. In the game Salimaki-Franssila, Finland 1999, there followed 10...2xb3 11.axb3 g6 12.éh1? (better 12.)d2 or 12.5\e1, with ideas like Wo 4 and/or f2-f4) 12...We7 13.1?! h5 14.43e2 g5, and now Black had the upper hand. 11.2h4 Wd7 12.5 2xb3?! 12...g8 was suggested as a possible improvement, but | think it’s already too late: 13.2h6! 7 14.Wt3 gxh6 15. Wrxt6+ Ge8 16.Wxh6 Ld8 17.6 +-; 12...87 13.14 4. 14...2ig8 15.25 Axf5 (15...Ag6 16.ah6 or 16.Afxg7+) 16.0xf6+, 14... df7 15.\xg7, 14...0-0 15.2h6 or: 14..c6 15.Wxe6+ Wxe6 16. Dxg7+ Sf7 17.2\xe6 Sxe6 Frick needed only seven minutes to reach this endgame with a clear pawn up - whereas the black clock showed 90 minutes. White converted his advantage convincingly. a 1-0/41.The Van Geet Attack 1.2)¢3 d5 2.e4 d4 3.\ce2 47 @ Sudden attack on the king After castling, Black's king is far from being safe. The c4-bishop plays an im- portant role in White's attack, e.g. when pinning f7: Schafer-Frey — Arndt Germany 1991 Black has just driven the c4-bishop to d5 by 16...a\ce5. Now he should de- activate this time bomb by 17...2e6!, as 18.2xb7 Eb8 19.2xa6 Exb2s can hardly be in White's intentions, e.g. 20.2b5 c4. Instead there followed the leisurely: 17...Ha7? 18.Axh5! An early ...h7-h5 is often tempting, while the disadvantages of ...h5 only show up later! More about this below in this chap- ter. 18...gxh5 19.2xe5+- Wxg5 19...\xe5 20.Wxh5 2e6 21.96 +-. 20.2d6 Ze3 21.2xf8 Dxd5 22.exd5 b6 23.We8 1-0 But even after exchanging the king's bishop Black should not wrongly believe that he is safe. In the following game, the bishop swap recently took place on a2. P. Bischoff — Fux Bern 1999 aon il BR 15.2xg7! Sxg7 15...2b4 16.\7h5. 16.054 16.xh6+!? Sxh6 (16...eh8 17.Dh5+) 17.2f5+ tg6 (17...2h7? 18.Wh3+ &g8 19.Wg3+) 18.Wg3+ Dg4 19.Wxg4+ Lt6 20.Wg7+ e6 21.Wh6+ f6 22.Wh3 had to be considered. But not least because of the offside position of the a2-rook Black retains chances for counter-play, e.g. after 22...ced7 23.2\xd4 f5. 16...@h8 17.2xh6 Probably 17.W4h3!? was stronger. 17..Hg8 18.Wh3 Dh7 19.14 exf4 .2xf4, and 20...2e5! should keep Black in the game. M. Larsson — Lavoisier corr. (1st North Sea team cup) 199848 Chapter II 16.Wg4! Wd7 Or 16...2h8 17.Ag6+. 16...Ag5 17-Axd6 (17.g3!7A14) 17... Wxd6 18.2xg5 fxg5 19.Wxg5 costs a pawn, 16...4f7+ was a less than ideal solution. 17.2)h6+ and 1-0 due to 18.Ag6+ hxg6 19.hxg6 gxh6 20.Exh6+ sg8 21.Wh3. The final example leads to the f2-f4 theme. However, here this is not played as a pawn lever but to make space for the rooks! Jo. Bastian — Luz Bastos corr. 1999 16.f4 exf4 17.2f3! After 17.05 Wxe5 18.23e4 Bfe8 19.ax16+ Wxf6 20.Wxh7+ &f8 Black escapes 17...2)d8 A...De6. 17...fxg3? 18.8xf6 +- and 17... De5? 18.2xi4 +— are out of the question; 17...Bfe8!? 18.Bafl Wt8 19.2\xf7! +. 18.2afl Ae6 19.2)h5! gxh5 20.293! Itis as if the black pawns did not exist. 20...fxg3 21.5xf6 “xgS 22.Wxg5+ GhB 23.Oxd6 cxd6 24.Wi6+ wg8 25.Wxd6 Bfd8 26.Wxg3+ L18 27.2xh5 with a winning position, 1-0/35. See, besides other examples in the cur- rent chapter, particularly Game 58 for a piece attack with this very pawn structure! @ The thematic lever f2-f4 Game 12 Schlindwein — S. Férster Germany (Youth-ch U17) 1991 1.03 d5 2.e4 d4 3.A\ce2 e5 4.293 2d6 5.204 De7!? Concerning other possibilities see Game 18. 6.d3 0-0 7.\1e2!? 7.23. 7...2g6 8.0-0 Yh8?! The c4-bishop makes Black nervous. Ex- changing it with ...We8 and ...2e6 is by far not sufficient for equality - White can point to the potential outpost f5, the bad d6-bishop as well as the lever c2-c3. Concerning the question, how bad (or not!) the doubled pawns are after 8...2e6!? 9.2xe6 fxe6 are, see Game 18. In the current situation, the g6-knight would be perfectly placed, while the ‘bad’ bishop pro- tects everything, e.g. 10.63 (10.242 “\f4! looks too slow) 10...c5 11.Wb3 (11.b4!?) 11...d7 - White has only a tiny edge: 9.74 exf4 9...f6 10.f5 would be a dream-like King’s Indian constellation - the king's bishop takes part in the attack, instead of idling on g2 (or g7 respectively). However, 9...2.g4!? 10.15 Df4 11.2xt4 exf4 12.4h1! was worth considering.The Van Geet Attack 1.203 d5 2.e4 d4 3.\ce2 49 10.Axt4 DxE4 11.2xf4 2xf4 12.Exf4 2e6 13.Wh5 Dunnington mentions 13.2xe6 fxe6 14. Exf8+ Wxi8 15.Wg4. White has De2-f4 in reserve, but attacking with Wh5 is much more natural. (bk Jae Gs a Ce gS I a OAS 13...2.x047! Thus Black increases his problems (x{7). A natural defensive move was 13...d7, and now 14.Baft Af6 (14...2xc4 15. dxc4 (6 16.2h4 h6 17.4f5+-). If now 15.Wh4 with kingside pressure or 15.%b5 with some ideas against b7 seem too vague, then the exchange sacrifice 15.2xf6 gxf6 16.Wh6 (A@h5) could be interesting: 16...2xc4 17.515 Hg8 18.0h5 Hg7 19.05 Wg8 20.Wxt6 eb 21.e7 Wi8 22.Exh7+! dxh7 23.Wh4# or 19...W8 20.Wxf6 2e6 21,2e7! with the decisive threat 2xh7+!. 1 g8 looks forced, but after 17.Exf6 White has at least compensation. 14.dxc4 Wes 14,8 15.Bh4 h6 16.15 or 14...16 15.Bh4 h6 16.5 with decisive threats 15.5h4 h6 16.415 Wes 16...{6 17.)xd4 was the lesser evil. 17.Wg5! Wg6 18.Axh6! 1-0 due to 18...xg5 19.4xf7+ dg8 20. ©xg5 with two extra pawns. Some minor omissions caused this drastic defeat. Glowatzky — ‘Iriknight’ corr. 2003/04 We The f2-f4 push seems to be out of reach for the time being, but it easily decides the game within a few moves: 14,15 0-0 15.Wg3 g6 15...n5 16.Wxe5+-. 16.2.95 Dbd7 17.5xe7+ Wxe7 18.14 g7 18...Qh5 19.Wh4 with overwhelming pressure on the kingside. 19.fxe5 Dh5 20.2xe7 Axg3 21.212, and Black may resign If Black has not yet castled, the advance {2-f4 may cause tactical fireworks on the critical f7-square, similar to the King's Gambit. Black frequently goes to the dogs; I've chosen a comparatively com- plicated - but nice! - example: H. Steiner — Lied! Vienna 199150 Chapter I! 12.f4 2g4!? 13.0xf7!? We7 After 13...2\xf7 14.2xf7+ &xf7 15.fxe5+ S98 16.exd6 &xd1 17.oxd1 dxc3 18.bxc3 &xc3 19.Eb1 2d4+ (19...b6 20.De2) 20.%h1 b6 21.214 &f7 Black should have obtained equality. 14.Axh8 2xd1 15.5\xg6 Wt6? 15... Wd6 16.f5 00. 16.45 The f-pawn now becomes a real force. 16...2.g4 17.295 Wd6 18.16 18... 2.18 Better was 18...0-0-0 19.fxg7 Wxg6, after which 20.26! is strongest: 20...\Wxg7 21. Exh6 Eh8 (21.018 22.9xh5+) 22.05 &2xf5 23.Exh8+ Wxh8 24.exf5. With the material balance being roughly even, the new-born passed f-pawn as well as the bishop pair are factors in White's favour. 19.f7+ @xt7 19...6d7 20.Bf6 Wc7 21.Axf8+ +-. 20.2x17+ In the final stage, the king is attacked by the white pieces. 20...8d7 21.086 We7 22.0xh5?! Even stronger was 22..¢.d5!. 22...2xh5? After 22...d6 Black could play on for some time. 23.206+ Sd6 24.2.c8# 1-0 Basically, accomplishing f2-f4 does not guarantee an advantage in itself - it's just what one normally does: Roques — Balaian corr. 1990 14.f4 After this Black can exchange his bad bishop. But that’s no reason for White to refrain from f2-f4, as he has no other sensible plan. This should indicate how questionable the concept of good and bad bishops is here, See also ®. 14...exf4 15.2xf4 2xt4 16.0xt4 Ag6 17.52 De5 18.h3 Wh6 19.b3 Hac8 20.Wd2 We7 21.25 Wd7 22.a4 b6 Both parties have obtained a nice square for their knight. The position is equal. 23.Hafl a6 24.Wf4 b5 25.axb5 axb5 26.Wg3 Ee7?? Counter-balancing the slight pressure on the kingside, Black prepares the lever ...c5- 4. 26...&h8 would have kept the equi- librium, but after Black's blunder it's over. 27.Ah6+ Yh8 28.Yxe5 Hd8 29.Exf6 1-0The Van Geet Attack 1.\c3 d5 2.e4 d4 3.2\ce2 51 ‘As we have seen, Black's equalizing meas- ure is ...exf4, plus afterwards taking e5 under control. He should not allow the advance f4-f5. Regarding this, | would like to outline some differences to the King’s Indian: Me Ld Ld Lhd White doesn't take on f5 because after e4xf5 g6xf5! (White is missing the analo- gous move in the Van Geet Attack) he cannot keep the g7-bishop under control in the long run - ...e5-e4 is in the air. On the other hand, Black doesn't play ...f5xe4 (after xe4 White has a nice square on 4) but ...f5-f4. The kingside attack with ...g5-94 etc is strong but not superior to White's queenside attack with c4-c5 as the g7-bishop is only obstructing its own pieces on the g-file. With a bishop on c5 - this would correspond to the Van Geet Attack, with colours reversed - the attack would be much more vigorous: Mazzoleni — Briozzo Villa Ballester 1993 18... Wb8 Gritting his teeth, he allows f4-f5. After 18...exf4 19.2xf4 (19.2gf5!?) Black is not yet ready to fight for the vital con- trol over e5, e.g. 19...Ag4 20.e5 &b8 21.Agf5 Dxf5 22.Wxg4+. The resource 296 (e.g. after taking on f4 twice) is not available due to the pin of the b3-bishop. 19...32xf4 20.2xf4 g5!? looks faulty, but it isn't easy to refute: 21.Exf6! Wxfé 22. @Dhd5 Wed (22...Wo6 23.e5 or 22...Wh8 23.2c7, with very good compensation) 23.213! 294 (23...Wh8 24.05) 24.2\xe5 ‘Bxe2 25.493 (the point, the bishop is trap- ped) 25...2xd3 26.5xd3 Hac8 27.2f with an advantageous endgame for White. = concerning the fork ...g7-g5 after f2-f4, compare Game 21, note g2 to Black's 6th move, including the insert which follows it. 19.5 Gh8 20.h5 Dxh5? More stubborn was 20...eg8, after which White continues his attack with 21.0xf6 xf6 22.94. Black lacks counter-play but retains some chances for defence. 21.Wxh5 +- Ags 21...f6 22.2xh6. 22.46! g6 23.Axg6+ fxg6 24.Wxg6 Exf6 25.Ext6 Wd8 26.217 1-0 A possible disadvantage of f2-f4 had al- ready been mentioned: The weakening of the e3-square. In the following exam- ple this weakness turns out to be illusory: Casper — Mébius East Germany (ch) Plauen 198052 Chapter I! Black's position looks solid, and the sec- ond player continues with some plausible exchanges, hoping for further relief. It looks like the slight initiative resulting from f2-f4 should peter out. But exactly the opposite happens: 11...2e6 12.2xe6 fxe6 13.2h6 Wd6 14.Wg5 2xh6 Also after 14...We5!? 15.24 White re- tains the initiative. 15.Wxh6 eS 16.Af4 Dg4 17.Wg5 Des 18.e5! Wc6 19.2e4 Ef7 So not e3 but {6 is the crucial weakness. 19...Axf1 20.Wxe7 (20.Exf1!?) 20...2ae8 (20...817? 21,5%6+ Yg7 22.5\xe6+) 21. Wo5 Ae3 22.564 (22.Axg6 hxg6 23. Wxg6+ Gh 24.16 Wxg2+ 25.Wxg2 xg? 26.2xe8 De3=) 22..Ext6 (22... £97? 23.D6h5+ kg8/wh8 24.2xg6+-) 23.exf6 with advantage for White, Black's king's position remains threatened (e.g. after h2-h4-hS). Still this was Black's best chance. 20.23 On his e3 outpost, the black knight is only a spectator. 20...Baf8 21.2f6+ Yh8 22.5h3 4g7?! Black was lost anyway, e.g. 22...2:3f5 23.94 Ded 24.Wh6 or 22...3d5 23, Whe (A Wxh7+!) 23...Axf6 24.exf6 ALS 25.2\xg6+. 23.Qxh7 Sxh7 24.Wxe7! 1-0 @ Exchanging bishops - some thoughts about ‘good’ and ‘bad’ bishops As already indicated, the issue of ‘good’ and ‘bad’ bishops may be irritating. Tech- nically, White's light-squared bishop is ‘bad’ and his dark-squared bishop ‘good’, accordingly, Black's dark-squared bish- op (e.g. placed on dé) is ‘bad’ and his light-squared (e.g. on e6) is ‘good’ Particularly with regard to the white bish- ops, this categorization is problematic. The f1-bishop is played to c4 or b5, in front of its own pawn chain c2/d3/e4, which overrules the labelling as a bad one. On the contrary, the c1-bishop is often restricted by Black’s pawn chain, e.g. d4/e5/f6/g7 or even -g5 (see ©, the pawn wall). Often Black uses his ‘good’ c8-bishop only to neutralize White's king's bishop, while his ‘big pawn’ on d6 effectively works against White's f2-f4 lever and helps to stabilize Black's position. “Bad bishops protect good pawns” - Suba, cited after John Watson, Secrets of Modern Chess Strategy; p.159, | recommend the study of the relevant chapter. D. Blau - Guzek (1) corr. (preliminary World Cup) 1994The Van Geet Attack 1.2\c3 d5 2.e4 d4 3.A\ce2 53 ‘Good’ vs. ‘bad’ bishop; additionally Black is burdened with doubled pawns. Nevertheless it's nearly impossible for White to make progress, as the d2-bishop has no effect. The game ended peacefully after some uneventful moves - 24.2c1 Wd7 25.Wh4 Eg6 26.hh1 WS 27.He4 2d6 28.We1 Wi7 29.Wa5 Wb7 30.2d2 Agfé 31.094 Yh8 32.He1 Wo7 33.Wad Eb8 34.Hee4 bis 35.291 Wb6 36.261 HfS 37.He2 Hh7 38,Wa3 2e7 39.Bge4 %2-%2, The light-squared bishop. A bishop on o4 can be - as we have seen - a powerful attacking piece. Therefore, often Black makes efforts to exchange it, e.g. by ...We7 (or ...2d8) together with ...2e6. If f2-£4 e5xt4 has been played, ...2c8-e6 is possible without preparations, as the possibility of doubled pawns is averted. It Black plays ...£2e6 first - as in the main line 1.23 d5 2.e4 d4 3.2\ce2 e5 4.2\g3 £286 -, then White strives for the exchange: typically by &b5-a4-b3; other methods are b2-b3, &f1-c4 as well as 2f1-e2-g4. However, there is another characteristic aspect. After the seemingly unfavourable exchange of one’s good against the opponent's bad bishop, often the queen takes over control of the colour complex. A queen on e6 is typical: Ekebjeerg - Khrenov corr. (10th Ech) 1966/72 In most analogous cases the bishops had been exchanged on e6 (here it went some-what differently). With the queen on e6, Black typically has a satisfactory position: 16.b3 Hfd8 17.We2 Ed7 18.cae1 h6 19. h4 b5 20.h5 g5 21.2c1 Hes 22.Ah2 DeS 23.015 218 24.94 Axg4 25.8xg4 Dc6 26.He2 De5 27.Wh3 Gh7 28.2.d2 c4 29.2e1 2c5 30.bef2 Ef8 31.2d2 Eff7 32.Wg3 2f8 33.Eb1 c3 34.2e1 £a3 Actually, Black’s bishop is really ‘bad’ now, however, the (good!) white bishop is so much restricted by Black's pawns that it is virtually ineffective. Of course, this is due to Black’s space advantage. Obviously, White is very solid on the light squares and has an outpost of his own (f5), so that neither party has any real cause to play for a win: 35.He2 Hd8 36.Wh3 Wc8 37.Bd1 “\c6 38.293 a5 39.Wog4 a4 40.212 2-2 Although this item is chiefly dedicated to the bishops, | consider it worth mention- ing that usually the second player has the greater resources when both queens are contesting the key diagonal a2-g8:54 Chapter Ii N. Neumann - Prihs corr. (preliminary German ch) 1993-95 Oppici - Medancic Toscolano 1996 12.43 A logical or stubborn move would be 12. a3; Black may continue with 12...5c8, 12...d3 or 12...h5, Compare the following game between the very same players: 1.403 d5 2.e4 d4 3.4\ce2 e5 4.293 Re6 5.043 16 6.c3 c5 7.Wad+ Dd7 8. &c4 Wb6 9.0-0 Ae7 10.d3 g6 11.2d2 297 12.cxd4 &xc4 13.Wxc4 cxd4 14.a3!? Hc8 15.Wa2 Ac5 16.Efct Ec6 (16... Dxd3? 17.oxc8+ Axc8 18.Wed) 17.e1 Qb3 18.Exc6 Axc6 19.2d1 a5, N.Neumann-H.Schneider, Germany 1994. White has allowed his queen to be locked in at a2, but still it’s equal. 12..5a5 13.We2 Be8 14.Wd1 Aacé 15.2.d2 We6 So the black queen took over control of the a2-g8 diagonal. However, this hap- pened at the cost of development. Now White achieves f2-f4: 16.h3 g6 17.Ah2 2g7 18.14 ext4 19. -&xf4 with approximate equality (/2-/2/36). Let’s go back some steps. In the follow- ing game we see Black initiating the exchange of both bishops, in the long tun to the pleasure of both queens: 7.43 Somewhat stereotyped. In this concrete situation, where Black has covered his centre less reliably (...2.06 instead of ..05), 7.c3! was stronger. 7...2e6 8.2xe6 Wxe6 9.cxd4 is promising for White, since the pawn duo e/f may be- come powerful in the early middlegame. It Black refrains from ...2e6, ...Wd6 doesn’t make much sense. 7...2.06 8.2xe6 Wxe6 9.2.d2 | don't like this. Preferable was 9.0-0, 10.22 and after that, according to the circumstances, a2-a4, @d2-c4 and/or 2-f4. The answer 9...h5 would be pos- sible; however, after the exchange of the light-squared bishops an intended kingside initiative lacks effectiveness. 9...h5, Here this is played to exchange the dark-squared bishops by ...2h6, and not with the usual aggressive intentions. Therefore, the answer 10.h4 seems un- necessary, better was 10.0-0 and if 10... hd 11.022 16 (11...n3? 12.4\g5) - planning ...g5 with the pawn wall - so 12.c3!. Also 10.h3 was worth considering. 10.h4 2h6 11.Wet 16 12.a3 &xd2+ 13, Wxd2 ge7 14.a4 a5 15.0-0 2d8 16. Bfd1 Hd7 17.c3 b6 18.2\e2 0-0 19.cxd4 exd4 20.Wh6 Efd8The Van Geet Attack 1.\c3 d5 2.e4 d4 3.A\ce2 55 Both queens bustle about on the diag- onals of their exchanged ‘good’ bishops. Is this an argument to keep, in case of doubt, the ‘bad’ bishop on the board? Black has the somewhat more harmoni- ous position, however, the equilibrium is not seriously disturbed. A‘few words about the structural change after ...:2€6xb3 (or ...¢5xb3) a2xb3. Moser awards a slight edge to White in corresponding situations. And really, the semi-open a-file as well as the safe prevention of ...c5-c4 are seemingly attractive. But praxis has not been able to confirm this assessment; the paucity of examples tends to indicate a situation of stable equality. White may get in f2-f4, but if Black prepares for this in a sensible way, {2-14 will at best supply a temporary initiative which has to be supported urgently by other actions before it peters out. And here it turns out that White has deprived himself of queenside pawn levers in the diagram position. The following historic example demonstrates the necessity of a two-wing-strategy: Bogoljubow - Zubarev St. Petersburg 1925 ce * « a GSR AG / ne Yes, ood a old tee already adopted the Van Geet Attack, the first moves being 1.2\c3 d5 2.e4 d4 3.Qce2 e5 4.413 2d6 5.293 c5 6.204 Dcéb 7.03 206 8.445 2xc4 9.dxc4. Later on he obtained a typical initiative after f2-f4, but still the second player managed to hold his kingside together. Only the open- ing of a second front leads to success: 28.b4! He6?! 28...axb4 29.axb4 cxb4 30.@xb4! Wxb4? (30...He6 31.505 +) 31.Wxg6+ d7 32. Wxg8+-. 29.bxc5 bxc5 30.5b1+— 4 Bb5. 30...216 31.Wxg6+ td8 32.b5 Ad7 33.Wxh5 Bf6 34.3 We6 35.Wh8+ e7 36.21b7 We6 37.Wg7+ 1-0 Summarizing, it should be said that one should be wary of any dogmatic views regarding the exchange of bishops. Abstract thinking inevitably leads one astray; one should pursue concrete goals without wasting too much thoughts. ‘on ‘good’ or ‘bad’ bishops.56 Chapter II © The march of the h-Pawn: «h7-h5(-h4-h3) One of the most tested counter-play methods is the march of the h-pawn, in the opening as well as in the early mid- dlegame. Basically, White may * block the pawn by playing h2-h4 (weakening the g4-square) or * h2-h3 (not a weakening; at once re- stricting the c8-bishop, but the g3-knight has to retreat after ...n5-h4), or * ignore the threat. The opinions of the experts differ in many concrete situations; so it is all the more difficult to draw general conclusions. 6.h4 blocks (of course, 6.. 7 follows). 6.h3 (with &b5 or 32c4 a good solution!) is not advisable as after 6...h4 7.2e2 the f1-bishop is locked-in. 6.2.64 or 6.2b5 invites Black to push his pawn to h3. The g3-knight. After ...h4 it can go to e2, f1 or ht. The first case (e2) is the most trivial one, therefore | need not illustrate it. ht! is interesting. In my working data- base, | found a couple of dozen examples in which 4\g3-h1 was a good move - often voluntarily played, without a friendly invitation by ...h4. This is because the knight often is ineffective on g3 (e.g. after ...97-96) and is looking for a new basis of operation via h1-f2. For the motif ...h4 2g3-h1 there follows an extreme example, in which the first player buries his knights on h1 and h2 for a long time. Most cases are simpler, White soon plays 4\h1, f2-f4 (or f2-f3) and “Ah1-f2; see for instance Ekebjeerg- Porreca within Game 27, 5.25, line b). The blocking square g4 is an important factor if Black strives for a kingside attack. An important helper to conquer the g4-square, the h-pawn, is no longer available after ...h5-h4. W. Lautenbach — H. Jongman corr. 1995, Bernard Partridge mem AM gb 4 13.22 Perhaps a sacrifice on h3 was threaten- ing, also the g4-square is taken under control now. 13...2:ce7 14.c4 wbB 14...dxc3!? 15.bxc3 f5, as after a quick opening of the game the offside position of the white knights could have an effect. On the other hand, Black's army is also marking time. As Black was convincingly out-manoeuvred in the game, he should have tried this in any case. 15.b4 t4 The only alternative was 15...f5 16.ext5, but after that White’s blockade on e4 and g4 works too. 16.2x14 exf4 17.294 f5 18.exf5 2xf5 19.Wd2 g5 20.13 Black has got bogged down. His knight will never reach its dream square on e3.The Van Geet Attack 1.\c3 d5 2.e4 d4 3.2\ce2 57 - if he’s going via {5, the g4-bishop will capture it; and if Black plays ...&f5xg4, a white pawn is transferred to g4. 20...We6 21.Bfe1 Wg6 22.2 Ac6 23.b5 De7 24.4e5 2c8 25,We2 Wi 26.2xcB ®xc8 27.2hg4 Wg7 28.c5 b6 29.\e4 mS aa Z 7 The resurrection of the knights! White con- trols everything, the win is a mere formality: 29...2e7 30.24 Wg8 31.a5 Hd5 32. Deté Hxed 33.axe5 Wes 33...216 34.0d7+ and 35.Wed+. 34.Qed7+ 1-0 The answer 4\g3-f1. Obviously, f1 is usually only a stopping-off place, the knight moves on via d2-c4, h2-f3 or h2- 94. If Black's temporarily initiative peters out, the h4-pawn tends to be a weakness; often White is able to re-organize his troops with great force. van Geet - Beuger Haarlem 1992 Here White wins even without moving the f1-knight again: 10.211 We7 11.2d2 b6 12.813 2b7 13.WI2 Se7 14.03 Actually, White cannot take the h4-pawn right now (due to xh4? @xe4!), but this lasting threat is a burden for the second player, who lacks any prospects of his own. On the contrary, White may play ¢2-c3, e4-e5 and/or f4-f5 at an appropri- ate moment. Generally (not so much in this position) after ...h4 Qg3-f1, ¢2-c3 followed by ©\f1-e3 is an important idea too, namely if Black is forced to play d4xc3 or to re- capture on d4 with a piece. 14...0-0-0 It’s difficult to suggest other constructive moves. Now a new wound is opened up: 15.25! Edf8? On 15...2f8, 16.15 was very strong, besides White could first strengthen his position by 16.h2. 16.\xf7 Zh5 17.2.xe6+ &b8 Black has lost material, and his hopes for counter-play are rather vague. The f1-knight is on guard until the very end: 18.5 2d6 19.Axd6 Wxd6 20.2f4 DeS 21.94 Ehh8 22.Wh2 1-0 If one believes van Geet, then h2-h3 fol- lowed by g3-11 is in principle (if possible and if there are no concrete reasons against it) the best answer to ...h5. The phantom pawn on h3. On the one hand, on h3 the pawn tends even much more to be a weakness than on h4, On the other hand, it covers a sensitive square in the very neighbourhood of White's king, namely g2. The following extract illustrates the resulting dangers:58 Chapter I! Dieu — van Geet Mariac (thematic tournament) 2000 17...2.05+ 18.04 18.geh1 Axed. 18...Axe4! 19.dxc5 & Whi+ 21.012 Wxh2+ 22.043 2 23.Wel Axf1 24.Wxf1 g5 25.De2 Lxe2 (26. Wxe2 Wi4#) 0-1 Sure enough, this is more the exception than the rule. Generally, time is working against Black: Either he succeeds in caus- ing confusion in the enemy's camp (a nice example is Reefschlager-Glek within Game 27, 4.4, line b 4...2\c6); or the impact of ..-N5-h4-h3 falls flat, while White succeeds in a favourable regroup-ing of his forces Ekebjeerg — Popp Yj Uj TWO White is already slightly better, the march of the h-pawn remains ineffective: 15...n4 16.2e2 h3 17.93 DhS 18.205 Eb8 19.21 We7 20.f4 Simultaneously an example of the push f2-f4. White takes over the initiative. 20...exf4 21.0xf4 xt4 22.gxf4 hs 23.13 og7 And the energy in the position is released by the following liquidation: 24.05 2xe5 25.fxe5 Lxg5+ 26.2xg5 Waxg5+ 27.Wg3 Wxg3+ 28.hxg3 25 29. Ef4 2xd3 30.Exd4 25 31.h2+ (1-0/43) The final example illustrates a complex and unclear struggle as often occurs after ...h5-h4-h3: van Geet — van Scheltinga Amsterdam 1968 We7 12.b4 Firstly, undermining the queenside. 12...2.94 13.95 Dh6 14.43 2d7 15.f4 As in the previous example, {2-f4 is White’s kingside counter to ...h5-h4-h3. Still, as a result of ...n5-h4-h3 all Black’s pieces are rather well placed, while the white ones display a mixed picture: 2c4, 5 and Hf are heading towards the focal point f7. The e2-knight is, however, very unfavourably placed (e2-g3 is not possible due to the g3-pawn!) and at the same time hinders the queen from enter- ing the action, which also impedes the development of the at-rook.The Van Geet Attack 1.2\¢3 d5 2.e4 d4 3.\ce2 59 Therefore, mutual light and shade. Indeed, White succeeds in unearthing a forced sequence which completely de- stroys Black's centre (the consequence of c2-c3, b2-b4 and f2-f4), but there follows a counter-attack on the poorly placed a1-rook: 15...2.94 16.2xf7 Dxt7 17.2xf7+ Wxt7 18.fxe5 Wg7 19.exd6 dxc3 20.bxc5!? 20.295 was good enough. 20...62 21.Wxc2 Wxat After 22.Wb3? Hf8 23.814 Wg7 24.e5 Wi7 Black repelled the threats, finally 0-1/34. Instead of this, White had had several favourable possibilities, e.g. 22.23! BTS 23.24 Wxfl+ 24.coxft g5 25.2d5 gxf4 26.gxf4+ (Harding), A brief summary: Black fights for the initiative by means of ...h5-h4-h3, White fights back with c2-c3/b2-b4/f2-f4, though his position is still partly disorganised due to ...h3. play will be exciting and adventurous, with the outcome often de- pending on tactical details. The discussion will continue in some crucial theoretical lines, e.g. Games 17 and 28. In Game 17 the answer h2-h4, fixing mutual weaknesses on g4 and g5, will be scrutinised. However, the above mentioned maxim that time is working for White doesn’t apply in the case of h2-h4, as White has compromised his Position at least as much as Black has. Until now, we have excluded a particular aspect of ...h5, namely that this move forms the foundation of the ‘pawn wall’: @® The black pawn wall on the kingside A method often preferred by strong play- ers: Black erects a kingside pawn wall. There are two versions of the wall: the ‘bathtub’ with 4e5 - Af6 - Ag6 - Ah5 as well as the ‘dark-squared triangle’ Ad4 - &e5 - Af6 - Ag5 - Ah4. Often White pre- vents the transition from the first to the second formation by playing h2-h4. The wall has several purposes: ¢ Assaults against f7 as well as other types of kingside piece attack are excluded « Especially White’s twirling knights (compare for instance Game 11) are limited in their scope. Steinitz already recommended the restriction of the opponent's knights by appropriate pawn moves. * The advance f2-f4 is prevented or at least impeded. * Also the ‘good’ 21 feels its sphere of action being restricted. * Black obtains a space advantage. * Later on, the pawn wall itself might turn into an attacking tool. Of course, all this happens at the cost of Black's development. But as the game Wyder-Kuiphof shows, Black can afford this. Much more important is the control of the position. Structure is more impor- tant than development - this maxim from the closed openings applies here as well. Computers won't learn that so fast, and in many correspondence games, even from master-level and above, one gets the impression that the second players - facing unfamiliar strategic problems - tend to rely on their clueless computers60 Chapter Il In fact, especially in the Van Geet Attack in countless correspondence games a white anti-computer strategy has tri- umphed! Back to our current subject. White does not have to be too much frightened off by the above arguments, as the pawn wall only works well if Black gets too much freedom of action, i.e. if White employed an inaccurate move order. In other cases White blows up the stonework by 12-4, ‘or he obtains - while Black is occupied by building up his for now defensive formation on the kingside - strong queenside play by c2-c3 and b2-b4. Ekebjeerg — Boey corr. (5th Ech) 1967-71 As well as covering e5, 5...f6 is simul- taneously the laying of the foundation ‘stone for the pawn wall. 6...96! 7.0-0 h5! 8.h4 Prevents the expansion of the wall by 8...N4 and 9...95. 8...247 9.d3 2d6 In similar situations even 9...95 10.2)hf5 h4 comes into consideration. in our case this would be too risky due to 11.5. 10.c3 Declares the queenside to be the battle- ground. You can study the resulting complex struggle in Game 21. The following example shows the ideal case for Black: Laurent - Michiels corr. 1980 2e6 5.213 Dd7 6.c3 c5 7.2b5 2d6 the weak 8.4h5? helped Black to establish a firm wall: 8...f8! 9.0-0 g6! 10.4g3 h5! 11.0xd4 cxd4 12.d3 (and now every pawn takes one step forward:) 12...h4 13.e2 {6 14.h3 g5. The odyssey of the e2-knight supplied the essential time for Black. In the diagram position, particularly f2-f4 is prevented once and for all: already the first preparatory step, namely g2-g3, fails to the incurable weakness of h3. In his helplessness White at least exchanges the strong e6-bishop: 15.b3 Ae7 16.204 2x4 17.bxe4 Ac5 18.8b1 Wd7 19.Wc2 And now stage two: the primary defensive wall mutates into a dangerous attacking tool. 19...94 20.hxg4 h3 21.4h2 hxg2 22.coxg2 Agé 23.2h1 Dh4+ 24.93The Van Geet Attack 1.2\c3 d5 2.e4 d4 3.ace2 61 ‘And now the inconspicuous f6-pawn has its say: 24.,.f5!! 25.63 25.exf5 e4+, 25.gxt5 Wg7+. 25...f44+ 26.c42 Axf3 27.ext3 Or 27.&.xf4!? @\xh2F. 27...0h3+ 28.%g2 Oxd3 29.5f1 We6 30.291 Wxe4 31.4b2 b6 32.5f2 we7 33.Axt4 exf4 34.5e2 He3 35.2.xe3 fxe3 36.Wxed+ xed 37.5b3 Ef8 38.dg2 &xh2 39.exh2 £6 40.v7g2 wed 41. Ha3 f7 42.ce1 Hf2+ 43.91 ot4 44. Btt Sg3 45.2xf2 Ax2 46.95 d3 47.41 ‘Sf3 48.Dxa7 e2+ 49.%e1 Ded 0-1 White showed maximal resistance, how- ever, he was already doomed in the first diagram position. One should also notice that the technically bad bishop - item © - didn't prevent Black from succeeding. The following game shows that Black, if he has a good grip on the position, can even neglect development for a while: Wyder — Kuiphof corr. (thematic) 1998 11...n5 12.h4 Limiting Black to the bathtub formation. 12...We6 13.b3?! The subject of @ - Black's queen now takes over control of the light squares. After 13.a4! b6 14.Wa2 c4, White could prevent this by 15.03 or 15.a5 first. 13...b5 14.We2 “\b6 15.4 bxc4 16.bxc4 Eb8 17.d3 Black's kingside is undeveloped, but this doesn't matter! The clear 7th rank guaran- tees the communication between the wings: 17...Eh7! 18.8b1 Shb7 With initiative on the queenside (+ or even +). The rest develops rather logically. 19.Ad2 a4 20.Exb7 Exb7 21.ab3 Ded 22.Wb2 Wa6 23.Wa3 Eb6 24.13 24.2xc5? Wxa3 25.2xa3 Ec6. 24...2)xa2 25.0xc5 2xc5 26.Wxc5 Ab4 27.Ed1 De7 28.e2 Wad 29.d2 Wb3 0-1 Now to White's counter-measures. As Black spends a lot of time on the pro- phylactic fortification of his kingside it’s obvious one must look for possibilities on the other wing: Manninen — Holmsten Helsinki (ch Finland) 199762 Chapter Il The small c2-c3 lever is already in action, now the big one follows with b2- b4. This is also an example of item ©: 15.b4!+ g4?! Anervous reaction, striving for counter-play. a) 15...0-0-07! 16.a\xd6+ Wxd6 17. bxc5 Wxc5 18.2a3+. b) 15...dxc3 16.2xc6!? (16.Axd6+) 16... &xc6 17.b5 or 16...bxc6 17.A\xd6+ Wxd6 18.bxc5 Wxc5 19.2.3 with a white edge. c) 15...cxb4 16.cxd4 exd4, and instead of the greedy 17.2xc6 2xc6 18.2\xd4 94 19. hxg4 h3oo now 17.4\xd6+ Wxd6 18.f4 looks good, though Black seems to hold on with 18...94. Possibly 16.2xc6 is more precise, but | want to stop here. 16.2\xd6+ Wxd6 17.bxc5 We7 17... Wxe5 18.243. 18.hxg4 2xg4 18...H3!? 19.13 27 20.cxd4 Axd4 21.A\xd4 Yxcd 21...exd4 22,243 with an enormous edge. 22.2xd7+ dxd7 23.2b2 The database gives 23.2d2 (probably an error in transcription while entering the game), but in this case Black could try 23...Wxd4+ followed by ...\Wxd3. 23...exd4 24.Wb3 De7 25.Wxb7+ 1-0 The final game of this passage shows how many concessions a grandmaster is ready to make just to erect the wall: van Geet - Bobotsov Wijk aan Zee 1968 13...H5 14.54 A typical answer to ...h5: ...h4 is hindered mechanically, 5 and g6 are targeted. 14...95 15.0hf5 ©xf5 16.0xf5 218 17. Qd6+ &xd6 18.4xd6+ Relying on the strength of the wall, the Bulgarian grandmaster Bobotsov accepted great risks: a position full of holes, part- ing with the bishop pair, allowing White's queen to enter the black camp. For the time being, White has the initiative. 18...Wd8 19.Wc5 We7 20.Wb6 Lh7 21.Bac1 According to van Geet, Black’s position would have “immediately collapsed” after 21.h4 gxh4 22.f4, | am not so sure about this (22...h3e), 21.2act is all right, White keeps control over the position. 21...We7 22.Wb3 218 23.Wb5 Wd6 24.2d5 3c7 25.Wa4 &g7 26.b4 2d7 27.bxa5?! After 27.b5 2\e7 28.2xc7 Wxc7 29.214 Black still had to defend carefully. 27...\e7 28.b3 xd5 29.exd5 29.Wxd5 Wxd5 30.exd5 Hac8 31.Exc7 &xc7 32.41 would have been sufficient for a draw. But | feel that both parties played for a win. 29...2ac8 30.0xc7 With 30.2c4! White could keep prospects for initiative. Van Geet loses this po- sition with bishops of opposite colours surprisingly fast, probably because he hunted after a will-o’-the-wisp in mutual time trouble: White aims for g7, but Black reaches g2 one move earlier. 30...2x¢7 31.:2b4 Wa6 32.h3?! Wb5 33. Wa3? Wxd5 34.218+ Yg6F 35.We7?? £06 0-1 @ The doubled c-pawns Often the question arises whether White should play his bishop to c4 or to b5. From c4 it eyes 7, after &b5, doubling Black's pawns on c6 is in the air. But is