botvinnik_move_by_move.pdf

C Lkw

Botvink mov

EVRYMAN CHS

 mov

' ,tf ry( hesco m

First pubishd  2013 by Goucstr Pbishrs Lmitd, Northbrgh Hous, 10 Northbrgh trt, Lodo ECV T Copyright © 2013 Cyrus Lakdawaa Th right of Cyrus Lakdawaa to b idtfid as th author of this work has b assrtd  accordac with th Copyrights, Dsigs ad Patts ct 1988.  rights rsrvd. No part of this pubicatio may b rproducd, stord  a rtriva systm or trasittd  ay form or by ay mas, ctroic, ctrostatic, agtic tap, photocopyg, rcordg or othrwis, without prior prmissio of th pubishr. Brtsh Lbra Catalogug--Publcato Data  cataog rcord for this book is avaab from th British Library.

Kd Kd IB N: 978-1-78194-104-1 Ebook IBN: 978-1-78194-103-4 Distributd  North mrica by Th Gob Pquot Prss, P.O Box 480, 246 Goos La, Guiford, CT 06437-0480.  othr sas quris shoud b dirctd to Evrya Chss, Northbrgh Hous, 10 Northbrgh trt, Lodo ECV T 020 7253 7253 7887 7887 fax: 020 020 7490 3708 3708 t: 020 ma: fo@vrymachss.com; wbsit: www.vryachss.com Evrym is th rgistrd trad mark of Rdom Hous Ic. d is usd  this work dr icc from Radom Radom Hous Hou s Ic. Ic. Evrym Chss ris Chif advisor: Byro Jacobs Cossiog ditor: Joh Ems ssistat ditor: ichard Paisr Typst ad ditd by First Rak Pbishg, Brighto. Covr dsig by Horatio Motvrd.

About the Author Cyru Lakdawaa is a Iteratioa Master, a former Natioa Ope ad America Ope Champio, ad a six-time State Champio. He has bee teachig chess for over 30 years, ad coaches some of the top juior payers  the US. o by the uthor: Play the London System A Ferocious Openin g Repertoire The Slav Move by Move 1    d6 Move by Move The Caro-Kann Move by Move The Four Kni g  g hts Move by Move Capablanca Move by Move The Modern Defence Move by Move Kramnik Move by Move The Colle Move by Move The Scandinavian Move by Move

Contents About the Author Author Itrod uctio 1 Botvk o the Attack 2 Botv B otvk k o Defece De fece Eemet 3 the Imbaaces 4 Botvk o 5 Botvk o 6 Botvk o Idex of

Bibliography 1 00 Selected Selected Games Mikha Botvik, (Dover 1960) Botinnik-Petrosian The 1963 World Chess Championship Match Mikhai Botvk Botvk (New  Chess Chess 20 10) Botinnik's Best Games 1947-1970 Mikha Mikha Botvik Botvik (Batsford (Batsf ord 1 972) Botinnik's Secret Games Ja Tmma Tmma (Hardge Simpoe Pubishig Pubishig 2006) Chess om Mohy to B owinnik Imre Koig (Boaa Books 1950) My 60 Memorable Games Bobby Fischer, Fischer, (Batsford 1969) 1969 ) My Great Predecessors Vol. , arry Kasparov (Everyma (Everyma  Chess 2003) Pachman's Decisive Games Ludek Pachma (Pitma Pubishg 1975) Tal-Botinnik 1960 Mikha Ta (Russe Eterprises 1970) Twelve Great Chess Players and their Best Games Irvg Cherev (Oxford Uiversity Press 1976)

Introduction "All told, there is not a sin gle weakness in his armour. 1  Reub Reube e F Fe. e. O August 17th 1911,  St Petersburg, a tita of the game etered the word. Mikha Botvk Botvk was bor to a detst mother, ad a father who was a deta tecicia. He eared chess at the ubeievaby ate age (for a word champio) of 12. It was ove at first sight. Botvk dspayed staggerg atura taet (athough he camed, rather outrageousy, that he had itte) ad, through the hep of his coach, Abram Mode, wo the 1931 USSR Champioship at age 20, the yougest to do so. I this period he casuay aexed a PhD  Eectrica Egeerg as we. I fact he cotue cotued d work as a eg egee eerr eve as word champio champio  uth uthkab kabe e by today's requrem r equremet etss to reach the the most exated eve. Botv Botvk k camed camed  a caim caim I dot beieve at a!  that his side job as egeer egeer actua actu ay y hepe heped d hm hm  his chess, sce he was aways hugry to pay. By 1936 he was perhaps the strogest payer  the word, demostrated by his performace at Nottgham, with a udefeated tie for first with Capabaca ad ahead of Word Champio Aekhe. Due to the terruptio of WWII, Botvik had to wait tweve og years before he became the officia sixth Word Champio, after havg wo the great 1948 Word Champioship touramet at The Hague/Moscow. He domated the evet, surgg a fu three (!) pots ahead of his cosest riva, Smysov. There were whspers that the Commuist Party authorities forced Botvik's Soviet rivas to throw games, but there s o proof of this. A simiar charge was made ater that Broste was forced to throw the ext to ast game  hs Word Champioship match versus Botvik, yet Brostes widow vehemety deied the caim ad said Botvik drew the match (ad retaed his tite) fair ad square. Max Euwe oted: "Most payers fee ucomfortabe  difficut positios, but Botvik seems to ejoy them! The match format, Botvik's forte, he cosidered the utimate test of oe's character. Botvik hed o to the tite, which he subcosciousy cosidered hs private property, for a fu 15 years, with two termissio termissioss  whe Smysov ad Ta briefy briefy "borrowed his tite tite.. Botvk's egthy reig quite possiby surpassed Lasker's, sce Lasker teded to dodge his great coteders, whereas Botvk faced a of them. Botvik, through dt of his superior preparatio methods, decsivey wo both rematches. Smysov he simpy outprepped ad outpayed strategicay. But perhaps most impressive was how he dodged Ta's fratic attempts to compicate ad forced hs youger, ess experieced (Word Champio!) oppoet to bocked positios ad edgs. Botvik quashed every attempt to cofuse, ad regaed the tite  covcg fashio, abeit bostered by Ta's  heath. Botvik  aog aog with Morphy, Morphy, Capabaca, Capabaca, Fscher, Karpov (ad (ad Carse! Carse! ?)  was the greatest strategst of his day (or ay day!). A argumet ca be made that Botv Bo tvik ik was the the sge sge most mos t importat importat chess figure of the the 20th cetury cetury  yes, you heard me correcty. Perhaps eve more so tha Fischer. The reaso: payers such as Capabaca, Aekhe, Ta, Fischer, Karpov, Kasparov ad Carse are merey isoated geiuses, a of whom produced beautifu games, yet oe revoutioed moder chess trag to a formuatio, a schoo. Botvik, o the other had, through hs tesey rigorous pre-game preparatio teciques, was the father of the Soviet Schoo of chess ad, by proxy, the father of a moder day professioa preparatio pre paratio ad ad coach coachg. g. Botv Bo tvk's k's secret (to Westeers) trag teciques may be the ma reaso the Soviets took soe cotro over the word champioship tite for the ext quarter cetury, whe oy the aomay of Bobby Fischer ripped it from

Soviet hads. The reaso we a so fraticay order ad study the atest opeg books is due to Botvk, who uderstood the deep importace of opeg theory ad pre-game preparatio. Oe seses from Botvk's pay, the residue of a rigorousy efficiet persoaity, uttery capabe of toeratg faiure  himsef. Ad whe he did fai (his osg matches versus Smysov ad Ta) he retured to the rematches with demoic resurgece, upedg the preteders to what he cosidered hs private kgdom: the tite of Word Chess Champio. H was a str a, who, from my prsoa 1977 sim mtg with him as a t, ackd affabity. (H sammd d scrwd  th pics wh h movd ad gard at yor trrfid, pmp-facd writr through thos scary cok-bott gasss of his, as a str prcipa wod to a dffict studt.) Botvk, a fog, dvout Comst Party mmbr, was a ma his prs mosty disikd ad distrustd, yt cod't hp but rspct. H was pro to ak outragous ovrstatmts o prcivd charactr faws of his rivas, ad yt, o sss, vr bothrd to podr ay particar dfcts  his ow. Through chss, this crdiby cofrotatioa prsoaity discovrd a ov mthod of divrtg his momta r aggrssio to th harss ram of th abstract.

Botvinnik' tye With Botv, thr mrgd a w sty of pay I ca  high d aggrssio, yt arisg from stratgic, ot soy tactica bass. To my d Vadiir Krak (Botv's studtys, ys, I kow: obody quats Krak to such a aggrssiv sty, but havg writt a book o hm, I dcar to you it's tru!) is Botvk's spritua chss so, who mbodis Botvk's powr chss  th prst. s Capabaca, kh, Krs ad othrs ard to thir dismay, Botv was ot a m to b trfd with  batts of cacatio powr, ad wh h szd th itiativ  spciay  his prim  his forts aways ros. Iitiativ was aways th prim focus as w s  this book ovr ad ovr aga, Botvk rjctg atria offrs f thy trfrd with his tiativ, th way a picky atr waks through a apptzg discot bfft  with a ary mpty pat.

Botvik caimed his great weakess was his abiity to spot combatios at critica juctures. But I harbour grave doubts about Botvk's sef-cofessed, purported weakess. Havg goe over most of hs games  preparatio for the book, I was staggered to dscover that Botvik virtuay ever missed a combatio  his prime  the mid 1930s to the mid 1950s . If Houdini saw it, Botvik saw it too. Hs aeged weakess bega to arise from the ate 1950s oward, whe Botvik was past hs prime (yet ubeievaby, sti word champio!). Botv, k Laskr bfor him, ctivatd a psychoogist's sight to ach of his rivas shortcomgs, ad dfty ad diaboicay wapoizd this drstdg ovr th board  his pr-gam prparatio. For stac, f h payd Krs, h wod try ad rach a positio whr it was bad for Krs to op th gam (.g. th whit sid of a Nizo-Idia, whr Botvk's sid had th bishop pair), ad yt Botvk kw Krs ovd op positios! If h payd Ta, h frustratd th Latvia's ov of tactics by boggg hm dow  bockd positios ad dgs, whr Botvk rigd. Covrsy, agast th sdat Ptrosia, Botvk wod jar hm by provokg  ary crisis ad opg th positio. I this fashio, Botvk fd away his oppot's qrks ad waksss for his ow futr rfrc.

King of the pening Botvk pmbd th dpths of th ary stags of th gam, drstadg ad dissctg his s th way a ovist's had is popatd with a cast of dozs of charactrs. Botv virtuay kpt his oppots  mta shacks, most brathg a rivd sigh f thy maagd to scap that phas of th gam. H drstood his opg systms ik o othr bfor hm.  tiaty ad dpy did h drstad th uacs, that v payrs such as Krs, Ta ad mysov somtms faid to mrg aiv from th opg stag. H was th first word champio try to wapoiz th opg phas of th gam, usg it as a whip, which had th ffct of cowg rvous oppots to mk thortica dodgs. Each ary crsh of a strog GM oppot cam across as a warg shot to postrity itsf.

Botvik, ike Aekhe before him ad Fischer after him, strove for perfectio  hs pre-game prep, with a work ethic borderg o faaticism. He exempified the sprit of mode professioasm  a aomay  his age  of a game which was the cosidered a hobby, a peasat teectua pastime,  which oe reied upo atura abiity. He ever payed bitz: "Yes, I have payed a bitz game oce, he said, "It was o a tra,  1929. He was aso vehemet  his sco for the memorzatio of opeg variatios without uderstadg: "Memorizatio of variatios coud be eve worse tha payg  a touramet without ookg  the

books at a! He was methodica, amost to the pot of predictabiity. He woud brg to each game a thermos of secret cotet to ourish hs bra. he hs cock ra, Botvik woud cacuate variatios  purey mathematica fashio ("If 23 Rxe6, the I have the trick 23 ... Kh7! etc). he  oppoets were o the move, Botvk worked schematicay, verbay forgg pas ad potetia futures. Botvik's opeg/ pre-game research produced a rich yied of ew uderstadg, brachig out  mutipe directios. Through his ubeievaby high eve of eruditio, Botvik gave direct theoretica chaege to the opeg ethos of hs time  a compedium of es, cudg the Frech Defece, Caro-Ka, G rfe d, Siciia Drago, Nimzo-Idia, ad may, may other es. I fact, he cotuay atered ad improved upo theory  whichever es he payed, aways at the forefrot of theory. He had a discocertg habit of radicay aterg og-hed assessmets, amost as a route occurrece, ad systematized opeg kowedge to ew, previousy uheard of eves. I for oe am gratefu to Botvik, sce those who ack the creativity to vet ourseves (e.g. your writer!), ca sti imitate giats before us, who bazed ew theory o a route basis. Here we see the 14-year-od Botvk disante a great word champion  a simtaneos game. We are rended of the wo rds from The Who's Acid Queen "Your boy won't be a boy no more; young, bt not a chid. Game  JRCapablaca-MBotvk Lengrad (sim) 1925 Queen's Gambit Declined 1 4 5 2 c4 e6 3 Nc3 N 4 Bg5 Nb7 5 e3 Bb4

Botvik had a ifeog pechat for meetg quee's paw opegs with . . . Bb4 ad ... c7-c, Ragoz-stye positios. He sidesteps the more soid Quees Gambit Deced es  ... c6 ad  ... Be7.

6 cxd5 Capabaca beat Edward Lasker from Back's side after 6 Nf3 c 7 Bd3 Qa 8 Qb3?

 ( ): Athogh White's ast move was a

buder, very few of us are awake to combatioa possibiities this

eary i the game. hat did the usuay hyper-aert Capa miss here?

 He missed the bizarre anomay 8 ... b5!!, wng ateria no atter how hite responds.

Istead, the game cotiued 8 ... Ne4? (the atura move but ot the best) 9 -!? (offerg materia for deveopmet) 9 . . . Nxg5 (Capa aways veered toward the simpe, avoidg the great compicatios arsg from 9 ... Bxc3 10 cxd5! which Houdini rates at eve) 10 Nxg5 cxd4 11 Nb5?! (hite shoud sac with 11 exd4! dxc4 12 Qxc4 Bxc3 13 Nxe6! fxe6 14 Qxe6+ Kd8 15 bxc3 with reasoabe attackg chaces for the piece) 11 ... Nc5 12 Qc2 Nxd3 13 Qxd3 a6 14 Nxd4 dxc4 15 Qxc4 Bd7 ad Capa wet o to out-tecique his oppoet from this pot  Ed.Lasker J. R.Capabaca, New York 191 5. 6   exdS 7 Qb3 The quee is vuerabe o b3, both to a future ... c7-c5-c4 (or d4xc5 Nxc5), ad ... Be 6. Today, 7 Nf3 ad 7 Bd3 are ormay payed at ths pot. 7   cS 8 dxcS QaS The quee pies o to the p with the route of a farmer decidg which of hs ufortuate chickes s to be toight's der. 9 Bxf6 Nxf6  O-O-O?





Overy optimistic. The white kg's couseors, fata advsors, whisper sweet promses of coquest to his ears, ad covce  to sig a uwise decaratio of war. Ths opportuistic decisio s't exacty bo of the precisio or ogic to which we are ormay accustomed from Capabaca. If you decide to embark o a adveture, be sure ot to ru to the waitg arms of a eemy! he the powerfu cogregate  a fixed ocatio, it makes for a temptg target  you are a assass. Capa auches a umoduated otio with, oe seses, mged msgivgs ad exuberace, aowg  kg to wader precipitousy far from the atura security of his ow side. Ideed, he vetures a agitated ad cumsy demostratio o the queeside, which soo gets drowed out  a barrage of back threats.

: This decision certany doesn't fit Capabanca's profie, does it?

 Agreed, bt sms exde their own socia mores. Capa, not beng cairvoyant, doesn't reaize the kid n front of him n the sm is destned to be a word champion. Compare this game to Botv's pendng of Keres n Game 25. At this pont Capa fais to acqire understandng of his risng isery ndex.

   11 Nf3 Be6 12 N d4 Rac8

Perhaps the wrog rook. I woud have payed the other oe to c8; i.e. 12 ... Rfc8! ad  13 c6 bxc6, whe the a8-rook is avaiabe for b8.

13 c6 Capa desperatey attempts t o bock the ope c-fie.

: Yes, bt at the cost of openg the b-fie! hodn't White

just pay for a edg with the simpe 13 Kbl Bxc3 14 Qxc3 - ?

 Capa oved endngs  bt not lost endngs, which he wod enter after 14 ... Qxc3 15 bxc3 Ne4! 16 Rc1 Nxf2 17 Rg Rxc5, when White's strategic woes contne to accrete ike a cheica company's effent, srreptitiosy dmped nto the oca river.

13  Bxc3 13 ... bxc6 ooks promisg as we. 14 Qxc3 Qxa2 1S Bd3 bxc6 16 Kc2! A itte simu cheapo, threateg Ral. 16  cS! 17 Nxe6 Not ow 17 Ral?? cxd4 ad ws. 17  Qa4+! 17 ... fxe6? aows a escape after 18 Ral d4! 19 Rxa2 dxc3 20 bxc3 Ng4 21 f3 Nxe3+ 22 Kcl ad White shoud be okay, despite beg a paw dow, sce he acquires targets o a7 ad e6. 18 b3 Qa2+ 19 Qb2 The quee abrupty decides to eave, absovg herse from a vovemet  the matter. White's chaces ook grim  the edg whe juxtaposed agast Back's, but there is o rea choice sce retag quees with 19 Kcl?? fxe6 eaves White's kg fatay exposed to the eemets. 19  Qxb2+ 20 Kxb2 fxe6

·



Uderstadig daws, the I/=f sig at the tai ed of a difficut mathematica equatio: White is competey busted. Not oy  he a paw dow, his kg remas terriby secure. The youg Botvk embarked o the fa assaut with great purpose, ad ever gave his egedary oppoet a speck of hope.

21 £3 Rc7 The immediate 21 ... c4! ooks a shade more accurate.

22 Ra1 22 e4! Rb8 (22 ... dxe4 is met by 23 Bc4!) 23 exd5 exd5 24 Kc2 was White's best defesive chace. 22  c4! Exceet judgmet. Botvik's saivatg remag pieces uxuriate  the taste of hutg dow a word champios kg. The attack ist over, despite the fact that quees have come off the board.

23 bxc4 dxc4 24 Bc2 Rb8+ 25 Kcl The kg oops aroud, the way a druk attempts to get out of a chair but keeps fag back to it. Whe surrouded by the courageous, a ma  ashamed if he

doest foow suit. Ufortuatey, 25 Kc3? waks to 25 . . . Nd5+ 26 Kd4 c3! (threateg ... Rb4+, foowed by ... Nxe3) 27 e4 Nf4 ad ow 28 Ke5 (28 g3? Rd8+ 29 Ke3 Ng2+ 30 Kf2 Rd2+ mates  a few moves) 28 . . . Nxg2 29 Kxe6 Rb2 is hopeess for White. 25 000 N d5 26 Re c3! 27 Ra3 Nb4! Threateg to capture o c2, foowed by . . . Rb2+ . 28 Re2 Rd8! Toyg with . . . Rd2 ideas. 29 e4 Rc6! The rook affects a humbe posture with a servie huch to get past the guards.

: Why not 29 ... Rd2 mmediatey?

 Even when bsted, Capa was aways aert to opportuties for ischief. In this case, destitte of defensive resources, White tries his hnd n a semi-swnde with 30 Rxc3! when he sti harbours some hope of surviva.

30 Re3 Botvik's attack, ow competey out of cotro, trasforms to a uaterabe property of ature, outside of White's cotro. Capa cotues to resist desperatey as we as fruitessy. 30 Rxa7 Rd2! aso ws.

 ( ): How did Botvk fish his great opponent off?

 Now Back's trick works.

30 000 Rd2! 31 Rexc3 31 Bb1 s met by the crushig 31 ... c2!, so the bshop fds himsef tied to the sacrificia atar. 31 000 Rxc2+! The pot: X-ray attack. 32 Rxc2 Rxc2+ 0-1 We are unaccstomed to a 14-year-od kid mahndng a reignng word champion n sch a aner.

cknowedgement Many thaks as aways to editors GM John Ems and Jonathan Tait for vignty ceang p your careess writer's nmeros go of-ps throghot the book. Thaks aso t o Nancy and Tm for proof readng and compter back-p. May Botv's ron ogic percoate nto deepened understndng for s a. Cyrs Lakdawaa, an Diego, Jy 2013

Capter One Botvnnk on te Attack When researchng Botvik's games I qicky reaized that the entre book cod easy be comprised soey of his nmeros attackng games. When a yong mn, Botvk, ke most of s, oved to attack. Bt n his case attacks were systematicay bt. His asterpiece aganst Capabanca above is a cear exampe. He wod frst accre appreciabe strategic gans, and oy ater attempt to cash ot with direct assats pon the opposng kng. Botvk's attacks were rarey desperado styeyo know the ones I mean: yo emit a hoarse war-cry and charge headong with the soe ntention of fictng as mch daage as possibe before one side sccmbs. Istead, Botv's opponents were sbjected to engthy nterrogation nder the harsh gare of the amp, and were graday broght ow, their ragged forces shiverng n ther tented encampments, awaitng the arriva of the nevitabe fna bow. Game  IRabovch-MBotvk R Championship, Moscow 1927 Dutch Dence 1 4 e6

Botvik was equay comfortabe  the Frech, Semi-Sav or Stoewa Dutch.

2 c4 £5 Dutch it is.

3 g3 N£6 4 Bg2 Be7

: In the tonewa, isn't it better for Back to pace

his bishop o d6, a more

aggressive square, which aso fights for e5?

 We, I pay tonewa, abeit say as White via a Coe move order, and do ke to pace my bishop on the correspondngy aggressive d3-sqare i possibe. Bt there are argments for postng the bishop on e7 as we. If Back posts his bishop on d6: 1. ite may pay a set-up ike b2-b3, Bb2

ad the Ne5. The  Back exchages with a d7-kight, he gets forked ad drops a piece. However,  his bshop is o e7  ths positio, the . . . Nxe5 is just fe for

Back, who maages to pug up the e5-hoe. 2. White may chaege the d6-post with somethig ike 4 . . . d5 5 Nf3 c6 6 0-0 Bd6 7 Bf4! . Agai, Back woud be better off with a bishop o e7.

5 Nc3 0-0 6 Nf3 d5 It's officia  a Stoewa.

: Is the tonewa a sound ne for Back?

 I don't care for it as Back, bt there is no accountng for persona tastes. My wfe Nancy sqeas n joy and caps her hnds n deight at the thoght of a trip to Disneyand, whe I view the same trip as a wf descent nto vgar comerciaism. Instead, 6 . . . d6 wod be a Cassica Dtch.

7 0-0 c6 8 Qc2 8 b3 Qe8 is aother set-up for White. 8  Qe8





Botvik's favourite maoeuvre. He pas to swig the quee over to h5,  the eighbourhood of White's kg.

9 Bf4 I am aways wary of payg this move o the White side versus a Stoewa, may because Stoewaers, eary aways pathoogica attackers, just ove to toss  a quick . . . g7-g5! ? to go for mate.

: We then, what wod yo sggest as an aternative pan for White?

 My tendency is to pay for a qick b2-b4; eg 9 Rb Qh5 10 b4 Nbd7 11 b5 as toh-M.Kjovic, ovakian Team Championship 1999. I wod be very nervos here as Back, snce the qeenside opens with aarmng rapidity.

9  Qh5 10 Radl

It ist too ate for my favourite pa: 10 Rab1 Ne4 11 b4 Nd7 12 b5 g5 13 Bc1!? (I woud go for 13 Bc7) 13 ... g4 14 Ne1 Nb6 15 c Nxc3 16 Qxc3 Nc4 17 bxc6 bxc6, J. Hammer-S.Haubro, Oso 2012, whe  sti prefer hite after 18 Nd3. 10  Nbd7 11 b3 11 Ng5 s met by 11 ... Ng4!, which forces hite to weake with 12 h4 Ndf6 13 f3 Nh6 with growig compicatios. 11  Ne4 12 Ne5 Ng5

Botvik ooks for troube o h3. 13 h4!? Risky. Some of us iduge i the trasgressive eed to disobey that which is cosidered awfu. Principle: Don't unnecessarily weaken your kin g's pawn ont.

: What other move does White have?

 I wod go for a more catios approach ke 13 Bxg5! (this move effectivey short crcits the back attack's nervos system) 13 . . . B xg5 14 Nxd7 Bxd7 15 e3 and I prefer White, who sowy expands on the qeenside.

: Aren't yo worried abot Back's bishop pair?

 Not here, for two reasons:

1. From my experiece, Back's attack sags cosideraby  Stoewa structures whe you remove his kights from the board. 2. The positio is cosed ad probaby wi rema so whe ite pays f2-f4 ad c4-c5,  which case, Back's bshops may be more of a iabiity tha advatage. 13  Ne4 14 Bf3 Qe8 15 Nxd7 ite is we advised to swap a few pieces. 15  Bxd7 16 Kg2 Now ay future . . . g7-g5 wi be met by h4xg5 ad Rh1, seizg the ewy opeed h-fie. 16  Bb4!

Provokg ite's ext move. 17 Bxe4?! ite cedes cotro over key ight squares. The humbe 17 Na4 ooks better. 17 000 fxe4 18 Rhl Qh5! Back meaces potetia exchage sacs o f4, as we as threats to chop the white kight ad swipe hite's e-paw. 19 f3? ite egeers hs ow ru with ths weakeg move. 19 000 Qg6! The vegefu quees eyes arrow  deiberate cacuatio. The dua threats are ... e4xf3+, wg the quee, ad . . . Rxf4.

20 Kfl Evadg both threats at the cost of pacg his kg o the ope f-fie. 20 000 e5! Powerfuy troducg the ight-squared bshop to ite's kg. 20 ... Rxf4! 21 gxf4 Qg3! was aso very strog. 21 dxe5? No better was 21 Bxe5? Qg4! 22 Bf4 Rxf4! 23 gxf4 exf3 ad the attack's eadg dicators a out perform expectatios. His best chace ay  21 h5 Bh3+ ! 22 Rxh3 Qe6 23 Rh2 exf4 24 g4 Rae8 with a admittedy awfu positio for hite.

 (  ): Work ot a wlg attack for Back.

 tep 1: Destroy White's ony functiona defender.

21  Rxf4! A shot which bows a gapg hoe  the white defesive barrier.

22 gxf4 Step 2: Iftrate deep to the recesses of White's er sactum. 22  Qg3! Back threates both . . . e4-e3 ad . . . Bh3+ . The cotagio, gradua  its oset, ow proiferates o the dark squares. White's kg, exasperated by the trusive back quees edess mooogue, takes a deep breath, exhaes sowy, ad takes a sip of we, prayg his der-date ordea soo comes to a cocusio.

23 Nxe4 White's oy try. 23  dxe4 24 Rxd7 The f2-square appears to be the epicetre of diverse ambitios: Back's to egeer mate; White's to avoid that fate. Both 24 ... e3 ad 24 ... Bc5 ook ke they w for Back but this is a optica iusio. Oe ws, the other waks to a trap.

 (  ):

We must make a decisio. Choose carefuy.

 Back's qeen is the sper-vian nd the dark-sqared bishop is her  Me. 24  B!

The geeticay atered mutatio smes wardy at the chaege, which  reay o chaege at a, sce he ow possesses the stregth of eight me  oe body. The aterative 24 ... e3?? aso ooks devastatg, uti we otice 25 Rxg7+!. Kg ad quee trip graceessy over themseves ad wai ther oss  smutaeous, articuate woe. Back must resig, sce either capture of the rook oses to 26 Rgl.

25 e3 Qxf3+ 26 Qf2 26 Kgl Bxe3+ 27 K Qxf4+ ws the d7-rook. 26   Qxh1+ 27 Ke2 Qh3 28 f5 Qg4+ 29 Kd2 Rf8 ite's paws are goig owhere . 30 e6 Qxf5 31 Qxf5 Rxf5 32 Rxb7 Rf2+ 33 Ke1 Rf6 34 b4 Bxe3 35 Ke2 Bg1 36 e7 Kf7 37 e8Q+ Kxe8 38 Rxg7 ite pays o, much ike the poor ma who buys a ottery ticket, which essetiay buys hope, far more tha the fitesimay miute actua chace of wg ad strikig it rich. At this stage, hite's war-weary defeders are oy capabe of offerg toke resstace. 38   Rg6 39 Rxh7 Bd4





Back's passed e-paw soo forces its way through. 40 cS Rg2+ 41 Kf1 Rf2+ 42 Ke1 The dyg od kg rasps fa structios to hs o-existet, ow dead geera, hs voice soudg ike metaic scrapgs: "Avege me! 42  e3 0-1 Game  M Botv k-V Alatosev R Championship, Moscow 1931 King's Indian Defence 1 4 N 2 c4 g6 3 f3 Bg7 4 e4 6 5 Nc3 0-0 6 Be3 e5 7 Nge2

7  Nc6 I dot t this move works very we agast the Smisch.

: What do yo sggest?

 I ke Kasparov's treatment: 7 . . . c6 8 Qd2 Nbd7 9 Rd (9 d5 is probaby White's best here) 9 . . . a6 1 0 dxe5 Nxe5! 1 1 b3 b5 1 2 cxb5 axb5 13 Qxd6 Nfd7! 1 4 f4 b4!, when Back's pieces sddey got scary active, A.Karpov-G.Kasparov, Lnares 1993.

8 Qd2 Nd7 Back teds to get squeezed after this move. Istead: a) 8 ... a6 9 d5 Na5 10 Nc1 c 11 dxc6 Nxc6 12 Nb3 Be6 13 Rd1 ad White exerted pressure o d6, J.ammour Hasbu-H.Nakamura, US Chess League 2007. b) 8 ... exd4  probaby Back's best bet; e. g. 9 Nxd4 Nxd4 10 Bxd4 c6 11 Be2 Be6 12 0-0 Qa5 with oy a miima edge for White, V.hikevich-P.Eders, Budapest 1996.

9 d5 Ne7 10 g3 : Isn't this move a bit odd?

 It is, bt remember the Kng's Indian Defence was n its nfncy n the 1930s. At that tme everyone payed it oddy. Aso, Botv's move scores we for White. With the hndsight of 70+ years of theory, 10 g4 f5 11 h3! is probaby White's optia ne, as n A.Karpov-P.Vrostko, Canes (sim) 1998, when Back is deied norma KID kngside counterpay.

10  f5 11 Bg2 fxe4 12 fxe4

12  Nf6 M. Botvik-L. Szabo, Europea Team Champioship, Hamburg 1965, saw 12 . . . a6 13 h3 Rb8 14 Ba7! Ra8 15 Bf2!, cevery gaig a tempo.

: How did this constitte a tempo gan?

 The dfference is with his bishop on f2, White can caste kngside. After 15 ... h6 16 0-0 Nf6 17 Be3 7, White was ready for qeenside expnsion, whie Back acked kngside counterpay.

13 3 Hatg ... Ng4 ideas. 13  b6 is oy deays c4-c5, which White ca prepare with a future b2(b3)-b4.

14 b3 I those days everyoe payed KID with excruciatg sowess! No moder GM woud eve cosider ths move. 14  Kh8 : What is the pont of Back's ast move?

 He frees g8 for his knight. Bt fter that, who knows? Yo actay pt your fnger on Back's man probem: he has no constrctive method of improvng his position. Ceary, the openng has not gone we for Back, who chafes under the restrictions of the rng athority. At this pont Aatortsev mst have experienced that awf ntition one gets pon the reazation that the power of penances are sometmes not enogh to grant grace.

15 g4





Payed  Botvk's youger stye: he goes after Back's kg. A more positioa payer woud perhaps caste kgside ad pay for the queeside c4-c5 break. 15   Neg8 16 Ng3 Bd7 17 0-0-0 Whe gog through Botvk's games, I was struck by the fact that he eary aways wo the game  situatios of opposite wg castg. Botvik the pure strategist is a myth! 17 g5 Nh5 (17 ... Ne8 ooks eve worse, sce White evetuay forces through h3-h4-h5) 18 Nxh5 gxh5 was temptg too. 17  h6 Perhaps cotempatg ... Nh7, but the move aows White to ceave ope a passageway to Back's kg eve more quicky. 18 g5 hxg5?! Back has to try 18 . . . Nh7 19 h4 h5, whe White may switch pas ad expad o

the queeside uchaeged.





Ceary the white aie mothership rues this quadrat of space  triios upo triios of cubic mies  ad woe to those spacefarers who dare to trespass upo the caim.

 (  ): Fnd hite's optma attackng pn.

 Ram throgh the h-pawn.

19 h4! The disparate activity ratios of the two parties are pa to see. 19  Bg4? The ufortuate bshop is issued his frst ad ast commad: a suicide mssio.  . . . g4  h5   hxg6+ Kxg6 was Back's ast dsma hope of surviva, though eve here his game ooks irredeemabe  its heret wretchedess after  Rdfl, tedg Nf5 ext. 20 hxgS+ NhS 21 Nce2  Nxh5! gxh5  Bh3! was more effective, eimatg Back's oy reiabe defeder. 21   N e7 22 Rh4 Qd7 23 Rdh1 White's rooks, wearg ther Suday best, show up o the back kg's doorstep, the way Jehovah's Witesses appear o me, ready to covert me, despite the promet "No Soicitors! sig  my wdow. 23  Kg? Back's kg, who cautiousy borrowed a arge sum of moey from a mob coectio, ow stes to hs phoe messages: two saes cas; three wags; five outright threats.

 ( ): Back's ast move was a bunder n a ost position. His position destabiized nto a qiverng, geatnos bob, unabe to withstand the sightest pressure. How did Botvk expoit it?

 tep 1: Einate the defender of the ight sqares by swattng the pest aside with contemptos ease.

  A dark, udergroud brach of the goveret is required to do the dirty work, which,  doe  the suight, woud be iega. Now the defece spiters ad cracks ope.   Step 2: Trasfer his ow bishop to a deady diagoa.

25 B3 Qf3 Step 3: Trap Back's wayward quee.

26 Rfl

A move which cofirms the back quees darkest fears: she has o pace to ru. 26 000 Nxg3 27 Be6+ 1-0 The bishop coses his eyes and istens with rapt biss to the back kng's tormented screams. Game 4 MBotvk-MYuovch R Championship, Lenngrad 1933 G nfeld Defence 1 c4 Nf6 2 4 g6 3 Nc3 5 4 Nf3 Bg7 5 Qb3 c6





A soid but passive choice. The moder treatmet rus 5 . . . dxc4 6 Qxc4 0-0 7 e4 a6.

6 cxd5 Nxd5 6 . . . cxd5 7 Bg5 aso offers White a edge, as  M. Botvik-S. Fohr, AVRO Touramet, The Netherads 1938. Back's fiachetto does't work a that we  a Exchage Sav format, sce the bshop hits a wa o d4.

7 Bd2 Of course the more drect 7 e4 is possibe too. 7 000 0-0 8 e4 Nb6 8 . . . Nxc3 9 Bxc3 woud  a sese justify White's earier Bd2 move.

9 Rdl N8d7 Back might cosider easg hs cramped quarters via swaps with 9 ... Bg4 10 Be3 Bxf3 11 gxf3 N8d7. Here 11 ... e6 12 h4 h5 13 a4 Qc7 14 f4 N8d7 15 f5 exf5 16 exf5 gave White a dagerous attack, V.Aad-S.No, Bad Mergetheim (simu) 1993.

10 a4 Perhaps White shoud toss  10 Bg5! to prevet a freeg . . . e7-e5.

10  as : hodn't Back break with 10 . . . eS - ?

 I agree, he shod ndeed. After 11 dxeS (or 11 Be3 exd4 12 Nxd4 Qe7 13 a NcS 14 Qa3 Nbd7 and foowng ... Re8, Back begns to generate pay aganst the e4-pawn) 11 ... NxeS 12 NxeS BxeS 13 Bh6 Qe7 14 Bxf8 Qxf8, Back ay get enogh dark sqare pay to compensate the oss of the exchange.

11 Be3 Now it wi be difficut for Back to egieer either . . . e7-e5 or . . . c6-c5. 11   Qc7 12 Be2 Qd6 Itedig ... Qb4. 12 ... e5?? is o oger possibe, as after 13 dxe5 Back cat recapture because he woud ose the kight o b6. 13 Na2! Oh, yet you dot! 13   e6 14 0-0 h6 15 Rcl fS!?





Sesig a gradua squeeze, Back ashes out for couterpay.

16 Nc3

White ca ao cosider 16 e5! (grabbg space at the cost of hadg Back a hoe o d5) 16 . . . Qe7 17 Nc3 g5 18 Nel ! with a cear advatage. 16  Kh7 16 ... Qb4?? hags the e6-paw. 17 Rfdl fxe4? Back aows himsef to be tempted by the w of the white a-paw. Istead, he coud geerate much eeded pay after 17 . . . Qb4! 18 Qc2 Nc4. 18 Nxe4 Qb4?!

: Why dbios? The move ooks ike it

forces White to swap quees or drop a

paw.

 This move is ess a mistake than a symptom, a fare-p of a degenerative disease: underestiation of an opponent's attackng potentia. The issng of deands from a position of nferiority seems an unwise poicy. Back fais to sense the danger and the preise of his pan is fawed, thereby fatay contanatng his position. He shod sette for 18 ... Qe7.

19 Qc2 Qxa4 A chid (the a-paw) herits ot oy a paret's weath, but hs eemies too. I Game 8, Botvk aso gives away hs a-paw  his attackg masterpiece agast Capabaca.

20 b3 Qa3 Back soo pays for  avarice sce he is uabe to dea with a chroicay weak square, susceptibe to sudde assaut.

 (  ): First, yo mst discover which

square ais Back, the work out the

expoitatio.

 Botv constrcts his uverse throgh sheer force of w. The chroic weakness of g6 costs Back the game.

21 Nh4! The kight exhibits the uervg characteristic of poppg up where east expected. ite's forces, havg accumuated vast reserves of resetmet agast the back kg, decide it's payback time ad come after him from chaegg ages. 21  Qe7 A move which presages a rather dsma tomorrow for Back's kg, athough it does have the beefit of shorteg his sufferg.

: Can Back save himsef by offerng an exchange with 21 ... f5



?

 Your ne is argnay better; the trobe is hite isn't obiged to take the exchange. Instead, he can contne attackng by 22 g4! Rd5 23 Nc5 Nf8 24 Nxg6! Nxg6 25 h4! with a crshng attack.

22 N xg6! Kxg6 Back's most awfu suppositios have come to pass. Now society stratifies via strict caste distctios  Back's kg, of course, firmy paced  the "utouchabe category.

 ( ): An exposive move

is afoot. ite to pay ad force mate.

 Ignore a the temptng knight discoveries and come at Back with the bishop.

23 Bh5+!! 1-0 The bishop destroys his riva, wipng away the stan of his very existence. Now the dreadf attack fows unabated after 23 ... Kxh5 (on a mooness night, the back kng tmbes overboard; otherwise 23 ... Kh7 24 Nf6+ 8 25 Qh7 ate or 23 . . . Kf5 24 g4 mate) 24 Ng3+ 4 25 Qe4+ (the qeen resurfaces) 25 . . . f4 (the spite bock!) 26 Qxf4 mate. Game M Bo k-V Alatosev Lenngrad 1934 Queen's Gambit Declined 1 4 e6 2 c4 5 3 Nf3 Be7 4 Nc3 Nf6 5 Bg5 0-0 6 e3 a6

: What is the idea behnd this move?

 Back hopes for B d3?! from White. Then he pans . . . d5xc4, . . . b7-b5, . . . Bb7, . . . Nbd7 and .. . c7cS, with a hybrid Qeen's Gambit Accepted, bt with an extra move snce White took two moves, not one, to r ecaptre on c4.

:  how doe s White dea with this idea?

 Pease see the next move!

7 cxd5! Trasposg to a QGD Exchage variatio where Back's . . . a7-a6 may ot ecessariy come  hady for him. 7 000 exd5 : If this is the case, can Back then head

for a QGD Semi-Tarrasch formatio with 7

. . . Nxd5 - ?

 He can, bt I don't see how ... a7-a6 heps him n ny way n the emi-Tarrasch. For exampe, after 8 Bxe7 Qxe7 9 Rc1 Nxc3 10 xc3 c6, as n B.Jobava-Zhang Pengxiang, Dos Heranas (onne bitz) 2006, Back remans n a terriby passive position and everyone is eft wonderng why he payed ... a7-a6, which is absotey of no se to him n this position.

 Bd3 c6?! This just doest fee right ad Back's ... a7-a6 begs to stick out as a wasted move. Perhaps his best bet  to offer a isoai positio with 8 . . . Nbd7 9 Qc2 c.

9 Qc2 Nbd7 A book positio from the QGD Carsbad e, but with the possiby useess ... a7a6 tossed  for Back, rather tha the deveopg move 9 ... Re8. But back the, a7-a6 wast cosidered a waste of time.

: Why not?

 We, the oy known pan for White was to caste kngside and then pay for a mnority attack with Rab, b2-b4, a2-a4 and b4-b5, where Back's ... a7-a6 wod actay come n handy. Botvik's next move was a competey new concept for the tme.

10 g4!!





The rewards of ashig out ofte exceed the gratificatio of patiece, its obverse. 10 g4 was a theoretica ovety at the time, ad a powerfu oe, which virtuay refutes Back's pay. White threates Bxf6 foowed by g4-g5. The effect of this move, i 1934, was the equivaet of a ma proposig to his parets that he is cosiderg givg up his successfu medica practice to foow his dream of becomg a street mime. Time to buid a coaitio force to take dow Back's kg.

: Isn't this jst a standard idea? White pns

to caste queeside ad itiate opposite

wg attacks.

 The move is standard  tod. At the tme this game was payed, it was a radicay new concept. We tend to take sch ideas for granted, when n reaity we shod brng to nd the Isaac Newton qote abot  standng on the shoders of giants.

10 000 Nxg4?! Caveat emptor  buyer beware ! The kight, ceary dispayg bithe cotempt for prcipe taboos, happiy satches a paw ad dares to defy the most hoy of precepts: Don't open lines to your own kin g for the opposin g major pieces. 10 . . . g6 s a better try, though eve there White's attack progresse s aarmgy quicky after 11 h3 Re8 12 0-0-0 Nf8 13 Kb1 Be6 14 Bh6 Rc8 15 Ne5 c 16 f4 cxd4 17 exd4 b5 18 f5 ad White's attack was faster, J.Garcia Padro-I.Miadovic, Las Pamas 1994. 11 Bxh7+ Kh8 Back's kg rai at the fact that White's attackers fai to accord him the deferece hs high birth demads. His desperate sese of isoatio grows ad he reaizes he acks frieds, whie his eemy's aies grow by the day.

12 Bf4 Ndf6 The defeders move suggishy, as if o barbiturates. Back is ao uikey to

survive 12 . . . g6 13 Bxg6 fxg6 14 Qxg6 Rxf4 15 exf4 Nf8 16 Qh5+ Nh7 17 Rgl.

13 Bd3 The bishop scampers back to safety. 13  NhS!? The kight's faigs oy produce a avaache of further frustratios.

14 3 Ngf6 15 Be5 Ng8 The defeders brace for the ocomig wave, hopig to arm the perimeter.

16 0-0-0





A casua gace tes us White's attack is destied to arrive first. 16   Nh6 17 Rdgl Be 6 18 Qe2! Eyeig h5. The quee, rich with scor, fiay codesceds to gace i the back kg's directio. 18  BfS? A buder, but Back woudt have bee abe to save himsef after 18 ... Nf6 19 Ng5 either. Back just budered  a aready troubed positio. The geometry shifted ever so sighty, eavg a opeg ajar for White's forces to eter. The fimsy mask is stripped away ad the back kg o oger abe to cocea his presece.

 ( ): Now BotvU orders

the executio with pri faity. How

did he do it?

Answer:

Step 1: Lure Back's kight to f5. 19 Bxf5! Nxf5 Step 2: The crushig dscovered attack eaves Back's uorgaized pieces uabe to save themseves. Back's kgside gets siced apart ike a order of sashimi. 20 Nh4! 1-0





hite begs a pncer movement: two arms rakng n chips won n a big poker hand. Back's hapess knights are harshy remnded that not a of s cn be extraordnary, for then who remans to take on the thankess task of beng ordnary? Game  MBotvk-VChekhover Moscow 1935 Rti Opening 1 Nf3 5 2 c4 e6 3 b3

·



Botvik experimets with Rti's Opeg, rare for the time because the idea of cotroig the cetre from the wig was sti cosidered a eccetric otio, oy reserved for hypermode. 3 000 Nf6 4 Bb2 Be7 5 e3 0-0 This ie occasioay trasposes to QGD Tarrasch-stye isoai positios. For exampe,  . . . c 6 cxd5 exd5 7 d4, as  C. LakdawaaL.Sussma, Sa Diego (rapid) 2012.

6 Be2 6 g3 eads to a more commo set-up after 6 . . . c 7 Bg2 Nc6 8 0-0 b6 9 Nc3 Bb7 10 cxd5 Nxd5 11 Nxd5 Qxd5 12 d4 Rad8 13 Ne5 Qd6 14 dxc5! Qxc5 15 Qe2 Nxe5 16 Bxb7 with a edge for ite due to the bishop par, Botvik/Pougaevsky Keres/Prs, exhibitio game, Amsterdam 1966. 6 000 c6 A soid but passive reactio. Back sets up a Semi-Sav vs. Reversed Quees Idia formatio.

: How is that passive?

 I reach a position smar to Chekhover's bt with White n a Coe vs. Qeen's Indian. In that version my bishop is more aggressivey posted on d3. Here we see Chekhover's bishop on e7, where it fas to fight for contro over the eS-sqare. Both 6 ... cS nd 6 ... b6 are more active methods of chaengng White's openng.

7 0-0 Nbd7 8 Nc3 a6 : Why did Back toss this move n?

 Back probaby ntended a qeenside expnsion with . . . b7-bS.

9 Nd4?!

·



We, this move proves that Botvik was o true hypermoder!

: Doesn't this jst ose time? What is the pont anyway?

 Bo tv's mysteriosy dbios move mst be n response to some fictiona exigency  or at east one I can't fathom. And yes, it does ose tme. Botv hopes to provoke ... c6-c5, which ay be a goo d move for Back! His move ntro dces an unprecedented eve of sbtety  either that or it' s jst a weak move! My gess goes with the atter theory. 9 d4, 9 Rc1 and 9 Qc2 a ook ike better optios for White.

9  dxc4?!

: And now Back vountary cedes some contro

over the cetre by swappig his d

paw for Back's b-paw?

 We, this is 1935, so even strong GMs payed this way n the openg. Today, I thnk an average cb payer wod know not to ake Back's ast move and wod atoaticay go for the sperior 9 ... cS!, thng "Thaks for the tempo!

: Ho w is that a tempo gan? Back moved his c-pawn twice as we.

 Yes, bt when White retreats his kight to f3, he wi have moved his kight three tmes to reach a sqare he aready reached n one! Then 10 Nf3 b6 11 cxd5 exd5 12 d4 Bb7 reaches a potentia hangng pawns (or isoai) foration. The oy other option is 10 Nc2, bt there seems itte pont n pttng the kight there.

10 bxc4 Now White's paws ejoy greater cetra fuece . 10  Nc5?! More time wasted, sce the kight ater gets the boot from a d2-d4 push. I wsh this hypermoder experimet woud ed soo! Somehow the mishaded (by both sides!) opeg stage of ths game grimy remds me of how I payed opegs as a

kid, tryig ad faig  a msguided attempt to trasform myse to a waabe, poor mas Nimzowitsch. Back shoud go for 10 ... c, though after 11 3 b6 12 a4!, he ow has to worry about  backward b6-paw, ad a4-a is aso  the ar. If Back hats ths with 12 ... a?!, he creates a hoe o b for ite's kight. 11 f4! Hatg ... e6-e ad fay fightg for some cotro over the cetre. 11  Qc7 12 Nf3 Oce aga hatg ... e6-e, ad preparg for Qc2 ad d2-d4. 12  Rd8 12 . . . b! ? s more active but aso weakes the queeside.

13 Qc2 Ncd7 I advace of ite's ext move.

14 d4





Thak God. Back to cassica chess. Botvik's epiphay: he s't suited for the hypermoder ifestye! It's as  a mad scietst, reaizg he ist cut out for a ife of evi, uatura experimets, uexpectedy throws away his ab coat, puts o a busess suit ad tie, ad goes to work for a surace compay to ear a hoest ivg. 14  cS Chekhover apparety experieced the same epiphay!

15 Ne5 Botvik prefers to pay to his stregth ad eter a hagg paws formatio.

: What is the aternative?

 I thnk he issed an opportuty to enter a knd of sper Benoi after 15 d5! exd5 (15 ... Nb6 16 Ng5! is aso proisng for White, snce 16 ... h6? runs nto 17 dxe6 hxg5 18 exf7+ with a decisive attack) 16 cxd5 b5 17 Ng5! Nb6! (not 17 . . . b4?? 18 d 6! Qxd6 19 Rad Qb8 20 Nd5 Re8 21 Bc4 and White is wng) 18 e4, when 18 ... h6 cn be met by the prosng sac 19 Nxf7! Kxf7 20 e5! Kg8! (20 ... Nfxd5? 21 Qh7! is crshng) 21 d6 Bxd6 22 exf6 with a dangeros attack.

15  b6 16 Bd3 cxd4 17 exd4 Bb7 18 Qe2 18 Radl was more accurate. 18  Nf8 Uderstadaby, both sides miss the impossibe-to-fid, computer-geerated e 18 ... Nxe5!! 19 fxe5 Rxd4! 20 Nb5! (20 exf6?? oses o the spot to 20 ... Bc5! ad if 21 Khl Rh4 forces mate) 20 ... axb5 21 Bxd4 bxc4 22 Bxc4 (or 22 exf6 cxd3 23 Qg4 Bf8) 22 . . . Bc5 23 Bxc5 Qxc5+ 24 Khl Ne4 25 Bd3 Qxe5, whe Back gets more tha eough compesatio for the e xchage. 19 Ndl!





Dua purpose  by terweavg a pair of mor ideas, the effect s the creatio of a major pa: 1. Botvk protects d4. 2. Botvk trasfers his kight to the kgside, where he pas to auch a drect assaut. 19  Ra7 Hopg to reforce f7 via og distace. GM Ludek Pachma, who did't ike this move, writes: "Such uusua moves are oy good o the rarest of occasios, ad this is ot oe of them. He suggests the passive maoeuvre ... Bc6 ad ... Be8, coverg f7. The troube with Pachma's e is that ite simpy picks off the bshop o c6 with a cear advatage; i.e. 19 ... Bc6 20 Nxc6 Qxc6 21 Ne3, whe f5 is  the ar ad Back remas far from equaity. Istead, Houdini suggests simpy 19 ... Ng6. 20 Nf2! Itetio: Nh3 ad Ng5. The immediate 20 f5! ooks quite promisg too: 20 ... exf5 s met by 21 Ne3! ad if Back "cosoidates with 21 . . . g6? ! there foows 22 Bxf5!, whe Back dare ot accept the piece. 20   Qb8 21 Nh3 h6

·



Back's ast move was desiged to prevet Ng5. So ow what? The baket of Back's defece may ook safe, sug ad warm, but it's aso too short, uabe to cover the kg's freezg toes.

 (  ): How did Botvk contne his attack?

 Pay the move anyway! Perhaps Back's ... h7-h6 nterpreted the position too iteray. The expermenta scientfic mode for data gatherng  tria and erro r  is sed here. The sac jst loo right. An experienced GM wod probaby pay sch a move withot cacation.

22 Ng5! hxg5 No choice, sce f7 fas if he deces. 23 fxg5 N8d7? He had to try 23 . . . N6h7 24 Nxf7 Nxg5 25 Qh5 (threateg mate  oe) 25 . . . Bf3! (the oy move; 25 . . . Nxf7?? 26 Qxf7+ Kh8 27 d5! e5 28 Rf5! mates) 26 Rxf3 Nxf3+ 27 Qxf3 Rdd7 28 Ne5 (threateg a quee fitratio to f7) 28 ... Bd6 29 Nxd7 Bxh2+ 30 Khl Rxd7 31 R, whe hite retas a domatg positio.

 ( (»: White's position, fsh with proise,

crushg cotuatios. How woud you cotue?

fds access to two

 The f7-pawn is the backened brise.

24 Nxf7!  #: 24 Nxd7! Nxd7 25 xf7! ! is even stronger, snce 25 . . . Kxf7? 26 Qh5+ Kg8 27 Qh7+ Kf8 28 Qh8+ Kf7 29 g6+ Kf6 30 Qh4 is ate.

24  Kxf7 25 g6+?! Botvik goes astray! He msses the pretty e 25 Qh5+ ! Kg8 26 gxf6 Nxf6 27 Rxf6! Bxf6 28 Bg6! Qd6 (28 ... Kf8? 29 Ba3+! mates straight away) 29 Rfl Bc6 30 Qh7+ Kf8 31 Rxf6+ gxf6 32 Qh8+ Ke7 33 Qg7 mate. 25  Kg8? After  ast move Back's strugges fa to dissipate a iota of axiety aroud his efeebed kg. Chekhover misses a mpossiby hidde defece with 25 . . . Kf8! (25 . . . Ke8? ao to 26 Qxe6 Nf8 27 Qf7+ Kd7 28 Bf5+) 26 Qxe6 Ne5 ! 27 Qh3 (ot 27 dxe5? Bc5+ 28 Kh1, tedg 28 . . . Bc8?? 29 Rxf6+ ! ad ws  Pachma, because of 28 . . . Bxg2+ ! 29 Kxg2 Rxd3 ad suddey it s White, ot Back, who is  deep troube) 27 ... Nf3+! 28 Rxf3 (or 28 gxf3 Qf4) 28 ... Bxf3 29 Rfl (or 29 Qh8+ Ng8 30 Rfl Bf6) 29 ... Ke8 30 Rxf3 Bf8 ad White st has a strog attack, but there is othig immediatey cocusive. 26 Qxe6+ After the brief but jarrg spatio-tempora distortio of the ast coupe of moves, ow a s set aright oce aga ad White s w g. His quees fagrat, teasg coquetry begs to make Back's very proper kg distcty ucomfortabe. 26  Kh8 27 Qh3+ Kg8





It ist so easy to take am at the target, which for ow remas shrouded  defesive mist. It ooks ke Back has everythg covered. But does he?

 (  ): Fnd a pan to fare p the attack.

 Transfer the bishop to e6.

28 Bf5! Nf8 29 Be6+ Nxe6 30 Qxe6+ Kh8 31 Qh3+ Kg8

 ( ): Fear of the unown and raw opportusm

risk war with oe

aother. Woud you sac the exchage o f6?

 In a heartbeat. White mst destroy the defender of h7 to progress i his attack. Hdini assessment: +27.85!

32 Rxf6! Bxf6 33 Qh7+ Kf8 34 Re! A key attackig pricipe: Don't chase the enemy kin g. Instead, cut o escape routes. 34  Be5





Back's ast move is aki to tossig cos i a fouta, hopg that hs wishes may come to fruitio. Chekhover hopes to buy his oppoet off by givg up his quee. However, Botvk is oy terested  the deivery of mate. 35 Qh8+! Moica, oe of my od cats, woud ki a bird  the yard ad the brg it side as a gift for me, ot reaizg that, touched as I was, I pace o vaue o a dead brd. I the same way, Botvik deces Chekhover's gift ad settes for othig short of mate. White ws after the crass 35 Rxe5 Qxe5 (materia is a fite commodity  imited suppy; Back's quee, a bored, affuet woma, has o probems  ife, so she vets a few just to keep her exstece terestg) 36 Ba3+ as we, but doest

force mate, the way Botvik's move does. 35  Ke7 From this momet o, the back kg's assemby e grows mootoous, each remag repetitive motio aturay foowg the other. 36 Qxg7+! The high strug quee, keyed to vioece, predictaby uges for Back's kg with murderous tet. 36   Kd6 37 Qxe5+ Kd7 38 Qf5+ Kc6 39 d5+! Kc5 The kg ives  a uiverse devoid of joy or hope. 40 Ba3+ Kxc4

 ( ): Let's see i we can work this one ot n

our md's eye. hite to

pay ad force mate  three moves.

 Jst foow the remander of the game.

41 Qe4+ Kc3 The kg's spasmg musces, aready sore from the previous day's overuse, ow experiece yet aother uwated workout. 42 Bb4+! Kb2 I his soitude, the kg fds ampe time for refectio o a misspet youth.

43 Qbl l-0





Mate! Your writer is nexpicaby eft speechess by Botvk's staggerng attackng skis. Game 7 MBotvk-S Tarakower Nottngham 1936 Old Indian Defence  Nf3 N 2 c4 6 3 4 Nb7 4 g3 e5 5 Bg2 Be7





The soid but passive Od Idia, the Phiidor of quees paw opegs. 6 0-0 0-0 7 Nc3 c6 8 e4 Qc7 Paifuy passive.

: What wod yo sggest?

 A more modern treatment runs  ... Re 9 h3 a6 10 Be3 b5, as n A.Fonyh-V.Maaik, t Petersburg 2001, when Back fights back aganst White's space advantage.

9 3 : Why is h2-h3 payed n this knd of position?

 White wod ke to rest his bishop on e3, withot the anoyance of ... Ng4 from Back.

9  Re8 10 Be3 Nf8?! The kight has few prospects o g6 or e6. It s't too ate for 10 ... Bf8 11 Rc1 a 12 Qc2, L.Va Wey-A.Zapata, Ledyards 2006, ad ow 12 ... exd4 13 Nxd4 Nc with at east a attempt at couterpay.

11 R1 6 12 d5 Grabbg more queeside territory. 12  Bd7 13 Nd2 White is ready for b2-b4 ad a evetua c4-c cash. 13  g5?!





Some woud deem such a move impusive, whie others woud abe it stct. Tartakower's arousa to fury fas fat as a opeed ca of yesterday's Diet Coke. This tirade, curiousy at odds with his previousy meek pay, quicky backfres, subsidg to subdued resetmet. Soo a combatio of despodecy ad advaced age eaves back's kg three degrees more huched over tha he was o the previous move.

: What is Back's idea?

 Back hopes to hibit f2-f4, or perhaps provoke it. In any case, it ooks ke a rather shady strategic decision.

14 f4! A move which casts grave doubt upo the true potecy of Back's "couterattack. Occasioay, drect, brutsh cofrotatio yieds more tha patiece ad subterfuge to achieve oe's aims. Botvik, a ferocious kg huter  hs youth, immediatey stigates the process of puishmet for Back's rash . . . g7-g move.

: Can't White jst ignore it and contne his focs on the other side?

 This is a styistic pont. I wod gess most po sitiona payers, me ncded, w od be more comfortabe payng somethng ke the eqay strong 14 cS! dxcS 15 Nc4 with growng qeenside pressure.

14  gxf4 15 gxf4 Kg7? There are more fissures ad gapig hoes i Back's positio tha he cares to recko, adg him  a wretched positio without recompese. hite eed ot eve vest ay materia for it. Back had to try 15 . . . Ng6 ad pray.

16 fxe dxe 17 c

Back is busted o may eve ad Tartakower coud eve cotempate resigatio here. 17  cxd5 18 Nxd5 Qc6 19 Nc4 Ng6 20 Nd6 Be6 21 Nxe7 Eimatg Back's soe defeder of his ow porous dark squares. 21  Nxe7

 (  ): Is n exchange sac on £6 sound, or is it gong overboard?

 The exchange sac is crshng snce it scks Back's kng nto the vortex.

22 Rxf6! Kxf6

Back's kg pays the roe of uwig host to ite's war party, who fies portetousy to his positio.

23 Q5 Ng6 After a momet of sour refectio, the e8-rook pucks up his courage ad sacrifices himsef. It ever occurred to him that obody wats him! Back hopes to appease ite by requishig hs hard-eared ga, the exchage  a devaued currecy, uikey to be accepted. 24 Nf5! I this harsh, sociay Darwia society, a deformities are uttery rejected ad Back's kg s eft to die  the sow o f6. The kight, who radiates maice from f5, is far more powerfu tha Back's worthess rook, sce Back's kg is preveted from rug away via e7. 24  Rg8 24 . . . Rh8 is simpy met by 25 Bxh6.

25 Qx6 Threateg mate o the move. The od back kg passes hs days puctuated with spasms of pa, stemmg from memories he wishes he coud expuge from hs tormeted md. 25  Bxa2 26 Rdl Threateg Rd6+. Aso powerfu s 26 h4!, tedg Bg5+ foowed by Bh3. 26  Rad8 27 Qg5+ Ke6 The timidated kg gazes tety at his ow feet, refusg to meet the quees eyes.

28 Rxd8 £6 Zwischezug. This stroke of imaged good fortue has the effect of foodg the ow bissfuy deuded back kg's bra with high eves of seroto.

 ( ): Both white qeen and rook hang ... or do they?

 Bit by bit, Botvk disantes every conceivabe aspect of the defence's achievements, unti a that reans of its former stre is a hmiated, bshng opponent.

29 Rxg8! Nf4 29 ... fxg 30 Rxg6+ (this rook makes everyoe fee awkward ad ucomfortabe, the way a best ma, fush with acohoicay-geerated goodwi, deivers a off coour toast/joke to the bride ad groom about pre-marita sex at the weddig receptio) 30 . . . Kf7 31 Rxc6 bxc6 eaves White two pieces up.

30 Qg7 1-0 The poxy consteation of Back's forces isn't a pretty sight and I sggest we a proceed qicky to the next game! Game 8 MBok-JRCapablaca  Tournament, The Netherands 1938 Nimzo-Indian Defence 1 4 N 2 c4 e6 3 Nc3 Bb4 4 e3





Botvik was faithfu to his beoved Rubste Nimzo-Idia hs etire ife. He didt care to pay 4 Qc2, the Capabaca variatio, possiby due to the opeg's ame! 4  d5 The most commo respose ad beief at the time: estabish a cetra grip with your paws. Nimz 0 witsch's radica idea of cotrog the cetre from the wgs was sti  its facy ad distrusted by most strog payers of era. Today, Back teds to prefer more fexibe set-ups after 4 . . . 0-0, 4 . . . c or 4 . . . b6. 5 a3!? Botvik's Variatio of the Rubste Nimzo-Idia. Keep  md that, at the time, this was a reativey ew theory. 5  Bxc3+ Capa refuses to back dow from the upstart ad pay the safer  . . . Be7.

6 bxc3 c The heck with Nimzowitsch's ideas. Capa, aways a cassica payer, decides to fight for the cetre with his paws. Back's positio s quite satisfactory. His ead  deveopmet compesates for White's bishop pair.

7 cxd5 I a ope positio the bshop pair is the co of the ream, athough the move heps Back free his game as we. 7  exd5 8 Bd3 0-0 9 Ne2!

: Why not deveop the kight to f3 nstead?

 White pans a centrakngside expansion with f2-f3 and e3-e4. The kight is to be transferred to g3 t o hep ot n this case.

9  b6! Preparg to swap off ite's moster ight-squared bishop via a6. 10 0-0 Ba6 11 Bxa6?! Uder the assumptio that Back's kight wi be out of pay o a6. The immediate 11 f3! gives ite a better versio;  e. g. 11 ... Re8 12 Ng3 Bxd3 13 Qxd3 Nc6 14 Bb2 cxd4 15 cxd4 Na5 16 e4 Nc4 17 Bl b5 18 e5 Nd7 19 f4 ad I prefer ite, who brews a attack o the kgside, V.Korcoi-A.Le, Johaesburg 1979. 11   Nxa6 12 Bb2?! Kasparov writes: "Aas, pioeers are doomed to make mistakes. 

: Why did Botvk deveop his bishop nto a wa on b2?

 He mst have had a hidden idea behnd it, bt for the fe of me I don't understnd what. Fortunatey, it is an easiy repairabe naccuracy. For every dbios decision, Botvk compensates with two exceent ones ater n the game. A the same, it is better to eave the bishop uncotted with 12 Qd3.

12  Qd7 13 a4 13 Qd3?! Qa4! favours Back. 13  Rfe8?! A atura move is't aways the best oe  the positio.

: I don't see any other reasonabe move for Back. What do yo sggest?

 White pns a sow centra bid-p, cmnatng n a kngside attack. Therefore Back shod adhere to the prncipe: Counter in the centre when attacked on the wing. The ony way to create a centra distraction for White here is to pay 13 ... cxd4! 14 cxd4 fc!, ntendng ... Rc4, generatng sdden qeenside counterpay. More mportanty, White jst doesn't have the eisure to do what he wants n this ne.

14 Qd3 c4?!

: Why wod yo criticze this natura gan of a tempo?

 Capa shod consider beng a itte ess partisan n his unwaverng faith n his qeenside ambitions. His move vioates the prncipe statng: Don't close the centre when attacked on the wing.

Botvik respods to your questio: "T s a reay serious positioa buder. Back evidety assumed that White woud be uabe to advace the e-paw ater, ad Back's superiority woud te o the queeside. However, Back's superiority o the queeside happes to be of o great cosequece i ths case, ad the breakthrough e3-e4 proves ievitabe. Back shoud have coteted himsef with the modest defece 14 ... Qb7. By keepg the cetre fuid, Back woud the decrease White's chaces of a successfu kgside buid-up, due to Back's costat threat of cetra distractios.

15 Qc2 NhS A case o f retrograde ambitio, tedg . . . Nc6-a5-b3 ad evetuay . . . Qxa4. 16 Rae1 Nc6 17 Ng3 Na5?! Aowg White the e3-e4 break without chaege. The payers terpret the positio with radica, atipoda outooks. Back shoud strive to chaege e4 with 17 . . . Ne4!, which is met by the strage 18 Nhl ! f5 (the desperate fight for e4 cotues) 19 f3 Nf6 20 Ng3 Ne7 21 Ba3 g6, ad eve though Back's kgside dark squares have bee weakeed, his positio ooks fitey more defedabe tha what he got  the game. I assumed White had a cear advatage here, but after payg aroud with severa computers from this positio, I reaized it was quite difficut to pay e3e4 without aowg Back a sorts of couterpay based o the ewy vacated d5square.

18 £3 Nb3 The waderig kight is dested to rema a tourist o the queeside for the remader of the game.

19 e4 Qxa4





A successfu pirate requres the foowg factors to esure success: 1. A ship capabe of outrug the authorities. 2. Lighty defeded, cocetrated weath to puder. 3. A market to uoad the spois. It fee to me that Back's missg eemets are umbers 1 ad 3 o the ist. The probem is that whie he does deed w the paw, he is aso very ikey to get mated o the other wg! I essece, Back refuses to yied to safeguards ad burs dow a bridges. If two of ite's eemies (Back's quee ad b3-kight) go off to war  a far-off ad, does this ot beefit him? Evidety, Capa didt beieve the bibica quote about a rich ma beg uabe to eter the kgdom of heave. I reaity, Back is  deep troube ad his extra paw wo't be of much assstace to him.

: Does Back have anythng better at this stage?

 I don't thnk so, and he may as we ake the attempt and grab the pawn. ti, Capa's strategy resembes that of a -year-od bionare who sggests to his new 23-year-od bride that they ake ove. ometimes determnation to get the job done jst isn't enogh!

20 e5 N d7 21 Qf2! I order to avoid . . . Nc5 tricks, which aow Back's kight back to the fight. Ateativey, Vadimr God,  Shakmaty v SSSR suggested the terestg 21 Re2! to couter ... Nc5. The idea was to sac a kight o f5 o the comg ... g7-g6 ad . . . f7-f5. 21 000 g6?!

·



Neither Botvik or Kasparov criticed this, but to my mid it's the osg move. Capa geerousy hads Botvik a target to ope with f3-f4-f5.

: What do yo sggest?

1. Brg Back's wayward quee back to the defesive fod with 21 . . . Qc6 ad eave the kgside paws aoe. 2. Push mady o the queeside with ... a7-a ad . . . b6-b5-b4, hopg to uderme d4. 3. Pray Back doest get mated! I fidded aroud with Houdini with these ideas ad Back seemed to obta better surviva chaces tha  the actua game.   three-step defensive pn:

22 £4 ite's e-ad f-paws race forward, each oe attemptg to outdo the other  height. 22  f5 No choice, sce aowg f4-f5 woud be catastrophic. 23 exf6! he I showed ths game to studets, may of the ower-rated oes were surprsed that ite wigy gave up hs passed e-paw. hite must pry ope Back's kgside at a costs ad this is the oy way to do so. 23   Nxf6 24 f5 Remarkaby, Houdini correcty assesses this  hite's favour, showg just how far computers have progressed  ther abiity to evauate a difficut positio. 24  Rxel 24 . . . g5? fais to stem ite's tida wave after 25 Re6 Rf8 26 Qe3 h6 27 Qe5. 25 Rxel Re! Kasparov offers egthy ad covcg aayss o hite's wg attack after 25 . . . Rf8 26 Qf4! .

 (  ): Capa redces ateria as qicky as possibe.

How did Botvk ramp hs attack up to etha eves?

 Force the creation of another passed e-pawn.

26 Re6! Not 26 fxg6 hxg6! ad Back's kight remas immue o f6. 26  Rxe6 No choice, sce 26 . . . Kg7? oses to the shot 27 Rxf6! Kxf6 28 fxg6+ Kxg6 29 Qf5+ Kg7 30 Nh5+ Kh6 31 h4! Rg8 32 g4! (threateg mate  two) 32 . . . Qc6 33 B a3! whe the og-ridicued bishop has his say  court. White forces mate  three moves . 27 fxe6 Kg7 28 Qf4! With a ugy threat. 28  Qe8! 28 . . . Qa2? oses at oce.

 ( ): Capa defty avoided this

How?

positio, sce White forces mate here.

 29 Nf5+! gxf5 30 Qg5+ mates n three moves.

29 Qe5 Kasparov ad Houdini poit out a simper w begg with 29 Qc7+ !, but the we woud be deprived of Botvik's magificet cocudg combatio. 29  Qe7

The positio teeters o the precipice of immet vioece. Now foows a combatio of moumeta scae ad scope, oe of the greatest  chess history. It is quite evidet that Botvk evisioed the positio to move 41, the ed of the game.

 ( ): How did Bo tvk break down Capa's resistance?

 The mother of a defection sacrfices. The bishop, once jst a fant otne of a shadowy figure at the edge of the forest, sddey approaches.

30 Ba3!! Qxa3 : Can't Back jst decne the sac, sidng the qeen to e8?

 This aows White's qeen ftration to c7 with decisive effect: 30 ... Qe8 (the awed qeen can oy ook at the ground unti comanded by the bishop to meet his eyes) 31 Qc7+! Kg8 32 Be7! Kg7 (or 32 ... Ng4 33 Qd7 Qa8 34 Bd6 and ates) 33 Bxf6+ Kxf6 34 Qe5+ Ke7 35 Ne2!, threateng Nf4, Nxd5+ and wns; eg 35 . . . Qa4 36 Qc7+ Kxe6 37 Nf4+ Kf6 38 Nxd5+ Ke6 39 Qe5+ Kd7 40 Qe7+ Kc6 41 Qe8+ and 42 Qxa4.

31 Nh5+!





A secod bow pierces the defesive barrier. 31 ... gxh5 32 Qg5+ The quee ooms arge over the puy, chihuahuaesque f6-kight, who fa with check. The back kg reaizes with dismay that there s o safe have. ite's quee decares to her hated brother o g7: "I wear a crow ow  ergo, I, ad I aoe, rue! 32 ... Kf8 33 Qxf6+ Kg8 33 ... Ke8 34 Qf7+ mates ext move.

34 e7





The e-paw, a adept  voodoo, squeezes the do. Istataeousy, Back's kg doubes over  agoy. Botvik reates a story at ths pot. Euwe waked up to Capa ad asked how he was dog. Capa shrugged ad said oudy eough for Botvik to hear, somethig to the effect of: "ythig is possibe, impyg that the positio may be draw by perpetua check. Of course Botvik saw through the attempted co job ad had worked out the w to the fish e. Eve more staggerg tha the doube piece sac of the actua combatio is Botvk's cacuatio power. He foresaw that Back had o perpetua check. 34 ... Qc1+ 35 Kf2 The fa part of this famous game  the kg starts his og wak up the board.

35  Qc2+ 35 . . . Qd2+ 36 Kg3 Qxc3+ 37 Kh4 Qel + 38 Kxh5 Qe2+ 39 Kh4 Qe4+ 40 g4 doest ater aythig. 36 Kg3 Qd3+ Cap a's quee does a the heavy iftig, foowg Botvk's kg aroud. White aso escapes from the checks after 36 ... Qxc3+ 37 Kh4, as  the previous ote. 37 Kh4 Qe4+ 38 Kxh5!





The dupicitous kg is a master of payg oe eemy agast the other. The h5paw must be removed. Back's checks are dested to ed very shorty. 38  Qe2+ 39 Kh4 Qe4+ 40 g4 Now possibe sce Back's h5-paw has bee eimated. 40  Qe1 + 41 Kh5 1-0 The back quee's dramaticay tiresome gesticuatios fay come to a ed. She is deied her most fervet wish: perpetua check. hat a wondros thng when a once-hazy fantasy takes form nto acta reaity over the board. Try one of the greatest attackng asterpieces of the 20th century. It happened agan: your noray verbose writer fnds hisef uncharacteristicay rendered mte by sch attackng prowess! Game 9 ADeker-MBotvk  vs.  radio atch 1 emi-lav Defence  4 5 2 c4 e6 3 Nc3 c6 4 Nf3 N 5 Bg5

Bodess s't ecessariy a vrtue  every stace. Deker opts to chaege with a e agast its very fouder. 5  dxc4 5 . . . h6 s the Moscow Variatio, payed more ofte today. 6 e4 b5 7 e5 h6 8 Bh4 g5 9 Nxg5 hxg5 10 Bxg5 Nbd7 The startg positio of the Botvik Variatio. Luckiy for me, this ist a opeg book ad therefore I seakiy circumvet a discussio of ths totay cofusg e.

11 exf6 11 g3 is the other mai brach here. 11 000 Bb7 12 Be2?!

: This move isn't nora, is it?

 The move is naccurate, bt we mst remember that the Botv Variation was st n its nfancy n 1945, so nobody (except Botvk) understood this fact. The probem with deveopng the bishop to e2 is that Back gets strong pay aganst the g2-sqare, which may force White nto payng g2-g3 ater anyway.  theory evoved to the concsion that White shod match Back's qeenside fianchetto with a kngside fianchetto of his own, to keep his kng sfer. 12 g3 is noray payed today, when 12 ... Qb6 13 Bg2 0-0-0 14 0-0 Ne5, V.mysov-M.Botv, Word Championship (5th matchgame), Moscow 1954, was another key Botvk game from this ne. Or i 12 . . . cS then 13 d5.

: Isn't 13 . . . N b6 strong here?

 Dang, yo  cevery drew me nto the theory, despite my frantic attempts to dodge! Apparenty, Pogaevsky refted it with the stung idea 14 dxe6! ! Qxd + 15 xd Bxh 16 e7 a6 17 h4! Bh6 18 f4!, L.Pogaevsky-E.Torre, Moscow 1981.

: How on earth is this a reftation? Back is p a rook n the endng!

 And stands ceary worse, possiby bsted, despite Hdini frantic assertions to the contrary! His h8-rook w never see the ight of day.

12 000 Qb6 13 0-0 0-0-0

·



Batte es are estabished ad both sides pa to urse their super-majorities o respective wgs, as we as auch attacks agast opposg kgs.

14 a4 a4 b4 b4 15 Ne4 Ne4 : Why not gan a tempo on Back's qeen with 1 a - ?

  trap. Back simpy pays 1 ... Qc7 and White can resign, snce Back threatens both mate on h2 and White's hngng c3-kight. In fact, yo jst demonstrated exacty why White shod fianchetto!

15  cS! Back's dormat pieces sprg to ife, pushg ite o the defesive.

16 Q b 1 If you yo u eed e ed to make make a move ike ike this this  a wide wide ope positio positio, it s a very bad ba d sig sig about the geera heath of your game. game . 16  Qc7 Qc7 A vugar checkmate  oe move aways seems magicay to upft a payer's sprits!

: o is Botvk payng for a one-move cheapo?

 Not at a. He simp simpy y pro bes White' White'ss kngside kngside for weakness.

17 Ng3 Ng3 Perhaps Deker had to try the (admittedy disma) e 17 g3 cxd4 18 Bxc4 Nc5 19 f3, though eve here hite's game remas o the critica st after 19 ... Nxe4 20 fxe4 d3!, cearg the gl-a7 diagoa. 17  cxd4 18 Bxc4 Qc6! It's a itte creepy how Back's quee dotes o her pampered chidre o b7 ad f8. Ah, the sweet ectar of the vugar oe-move mate aga! From this pot hite ever gets a chace to rest. Botvik quicky adapts to the ew ecosystem, his pieces the predators at the top of the food cha.

19 £3 d3!





Back bursts forth i a dazzg starburst of piece activity. Aarmgy, the groove  White's defesive barrier grows wider by the move, dested to be siced  ha, by thrusts of a hadsaw to a og of firewood.

20 Q c l The oy move, sce 20 Rcl?? Qc5+ 21 Khl Qxg5 picks off a piece. White's quee eyes Back's truders with apprehesio, makg certa to rema a good distace ahead of the pursuers. Oe gets the ueasy feeg that White's ast eight moves were amedme amedmets ts  a a attempt to correct corre ct his his first tweve. tweve . 20 000 Bc5+ Troubes Troubes pour forth for White, White, as a s hs hs oppoet oppoet's 's pieces stream out. ou t. 21 Kh1!

: Why doesn't White simpy bock with his bishop?

 It's not so simpe. White's gard is p. He defty sidesteps a cheapo attempt, havng been around the bock before. If he enters your sggested ne, Back wns with a combnation: 21 Be3 d2! (attraction) 22 Qxd2 Ne5 (discovered attack) 23 Qf2, when White's unstabe strcture ooks dried ot and crmby, food eft too ong n the refrigerator.

 ( ): Back to pay and force the wn of heavy ateria.

 Overoad. 23 ... Ng4!! and the knight is untochabe snce 24 fxg4?? Bxe3 25 Qxe3 Qxg2 is mate. The qeen profits from the discord n White's camp, howng off with the spois.

21 000 Qd6! Threa Threate teg g . . . Qxg3. The The quee q uee haes haes deepy dee py,, rechargig rechargig her ugs ugs for the the beowed tirade which foows. So smooth are Botvk's best attacks, they seem amost as if fashioed by atura desig, more tha huma thought.

22 Qf4 22 Bh6 is is met by the chi chig gy y quiet move 22 . . . Rh7!, Rh7! , whe White has o defece defec e to the comg doubg aog the h-fie.





Deker was udoubtedy bewidered by the rapid tesity of  dowward spra. Despite White's best efforts, the kgside proves ugoverabe. Ay accoutat woud form you that White's defece is  the red, periousy cose to fi fig g for bakruptcy. bakrupt cy.

  ( (   ): ): How can Back

smash his way through to White's kg?

 Throgh the process of mpermnence, those n positions of priviege and power don't aways rean where they are n perpetuity. White's kng reaizes he forestas the onset of adness oy by distractng hmsef from the the awfu trth  his power, power , his fe itsef, is on the wane, wane, abot t o be surped by those who despise him, unaided by those who profess to ove hm. Now he is forced to enter the sarcophags whe st aive.

22   Rxh2+! Rxh2+! 23 Kxh2 Kxh2 Die rich or die poor. It doest matter. It is of o soace to ador your tomb with treasure. 23  Rh8+ 24 Qh4 Rxh4+ 25 Bxh4





The persprg bishop arrives o h4  the ick of time, or so he beieves,  the aid of  kg. It ooks ike White is okay (he most certay st!). After a, he got two rook rookss for the the quee quee ... . .

  ( (   ): ): Or did he?

Back's ext move tears away the fimsy vei.

 Dobe attack. Back's qeen does ot pushment to the deservng. The rich get richer! bemon White's White's bishops, a t the qeen qeen's 's endess greed . 25  Qf4! 0-1 Game Game 0 STarakower-MBotvk taunton Memoria, Grongen 16 rench Defence 1 e4 e6 2 4 5 3 ex5

:  tad cowardy?

  wise decision. urvivng the openg aganst Botvk was no simpe task, so Tartakower takes a suitabe precation and veers cear of Botv's deady openg arsena. ometmes it is okay to give give away the first move advantage advantage f your y our opponent opponent represent repre sentss a ibrary ibrary of theory which which yo coud never

hope to match. Tartakower wod most certany have been kied had he entered the ntricacies of amost any other French ne aganst the maestro. Botvk said Tartakower payed this way "n order to excde any srprises n the openg ... "

3 000 exdS 4 Nf3 Bd6 S c4 I t this ie s more promisg tha 5 Bd3 Ne7 6 Bg5 Nbc6 7 c3 Bg4 8 Nbd2 Qd7, as  Z.Bogut-A.Grschuk, Europea Cup, Kaithea 2008. I reached this positio a few times as Back ad have the feeg he aready stads a shade better sce he ca caste queeside ad auch a swift kgside attack. S 000 Nf6

6 c Logica, sce it comes with tempo.

: Can White aso pay n pre pre isoai fashion fashion with 6 Nc3 0-0 7 cxd5 - ?

 This is possibe too, bt White gets a more passive version than norma after 7 ... Re8+ 8 Be2 Nbd7 9 0-0 h6 10 Bc4 Nb6 11 Bb3 Bg4 12 h3 Bh5 13 Be3 Qd7 14 a4 a 15 Rc1 Qf5! Qf5! 16 Bc2 Bxf3 17 Bxf5 Bxf5 Bxd 18 fxd Nfxd5. Yo are witness to a irace. Yor noray bmbng writer anaged to achieve the sighty sperior endng verss a Word Champion (and wn!), G.Kasparov-C.Lakdawaa, Internet (bitz) 1998.

6 000 Be7 7 Bd3 b6 8 cxb6 axb6 9 0-0 0-0 10 Nc3 Possib Poss iby y accurate accurate as Back's g4-p ow grows grow s irritatg. irritatg. I woud wou d pay 10 h3 h 3 as  G.Russek Libi-A.Vae, Sao Pauo 1991. 10 000 Bg4 11 h3 BhS 12 g4!?

·



Oe ma's territoria gai  aother's overextesio.

: Why does White agree to weaken his position to this degree?

 He mst, snce the aternative is the naseatngy passive pn-breaker 12 Be2. 12  Bg6 13 Ne5 Bx3 14 Qx3 c6 15 Bg5?

 ( ): The ever-vigiant Botv found

a trick to expoit ite's

atura atura but weak ast move . How?

engagng an excesior theme combnation.  By engagng

: What is an excesior theme?

 One of those:  chop on g4, then he takes e7, then I take his kight on eS, then he takes my

qeen on d8 ...  deas.

15  Nxg4! xg4! From researchig this book, I was staggered to dscover that from about 1936 to aroud 1950 Botvk Botvk amost ever missed a combatio  hs games. I had Houdini rug, ookg at hudreds ad hudreds of games, ad wheever Houdini souded the combatio aert, Botvik,  stereo, aways seemed to fd the same combatio as we.

16 Nxc6 If 16 hxg4 Bxg5 ad Back ws a paw; as does the e 16 Bxe7 Nxe5 17 Bxd8 Nxd3 18 Bx B xb6 Nd7 19 Bc7 Nxb2. 16  Nxc Nxc66 Tartakower Tartakowe r retas retas the materia baace baace at the cost of his his kg's kg's security. se curity. 17 Bxe7 Nxe7 18 hxg4 hxg4 fS! fS!

A great upeasatess for White. Botvk Botvk immediatey goes after White's weakeed kg. 19 Rael! White hads over the paw, reyg o his ead  deveopmet. Istead, 19 g5?! f4! ooks pretty scary for White; as does 19 gxf5?! Nxf5 with a strog attack. 19   fxg4 fxg4 20 ReS Rf3 Rf3 21 QbS QbS Ng6!? The addict sef-medicates, offerg yet more materia to brg o the bessed reief which attack provides. Botvik, who gave himse a excam for this move, sacs the paw back to crease his piece activity. His decsio is based o the theory that eective, partia defeat is actuay a key compoet to victory. Oe must frst ear to give up somethig to make gas ater o. I'm ot as cofidet; perhaps Back ca pay pay the more crcumspect crcumspe ct 21 . . . Kh8 ad ad try to hag hag o to the the materia.

contradict Bot vk's vk's assessmen assessmentt of the positi po sition? on? : Yo dare t o contradict

 We, normay I wod be too gtess to do so i reyng pon my own stunted bran, bt with 3200-rated big brother Hdini hepng ot, yes, I dare.

22 Rxd5!?

: Why didn't White force qeens off the board?

 I had the same thoght, bt then compter anaysis proved that White reans n deep trobe there as we: 22 Qxd5+ Qxd5 23 xd5 Raf8 (threatenng 24 ... g3) 24 Ne4 Nf4 25 Rb5 h5! 26 xb6 h4, nd ... g4-g3 ooms agan, a menace pon the andscape. I'm not so sre White can save himsef after 27 Rb3 xb3 28 axb3 Ne2+ 29 Kg2 Nxd4.

22  Qf6 23 Rg5?! ite's best chace to save himsef is to cetrae mady with 23 Ne4 Qf4 24 Re! (threateg Qd+) 24 . . . Rf8 25 Re8! . 23  Rf8 24 Ne4 Qf4 reateg ... Rh3 ad ... Qh2 mate. However, the seemgy hepess white kg merey feigs seep ad  ready to ca o his guards. 25 Qd5+ Kh8 26 Rh5 Coverg the matg threat, whie threateg Ng hie. 26  Rh3 reateg mate  oe. Botvk was  severe time troube at this pot.

27 Rx3 3 28 N g3 ite avoids the trap 28 Ng?? Qg4+ 29  Ne7!, overoadg his quee, which ca't cover both the g2-mate threat ad the g-kight. 28   Nh4 29 Qe4 Nf3+

·



ite's kig ca oy faux-smie at the itrusive guest, as he resigs himsef to the ew reaity. His firmity o f3 is destied to be a fixture  his ife ad have a parayzg effect for the remader of the game. 30 Khl Qxe4?! A misguided overture. Now his pas to checkmate beg to degeerate to a somewhat coheret facsimie of a earier tet.

: hodn't Back keep qeens on the board?

 trangey enogh, Bo tv doesn't coment on the move, bt yes, the compters scream to retan the qeens and cam that Back's decisive advantage sips after 30 . . . Qxe4. Perhaps Botv 's decision was based o n his tme pressre. 30 ... Qf7! was mch stronger, threatenng to fork o n d2. Hdini assesses it at -2.20, wng for Back.

31 Nxe4 Rf4 32 R1 6 32 . . . g 5 is sighty more accurate ad may ater save a tempo. 33 Rc3 g5?! Time troube, that maigat momet of each game where eve ubeievaby strog payers, without wag, degeerate to fustered begers. Ah, yes, the tyray of the improbabe. I our chess ives, how may good positios have we a tossed aside to chase some iusio? Botvik's chroic time probems eary mess thigs up. Back  sti wg after the obvious 33 . . . Nxd4, eimatg hite's dagerous, passed d-paw. 34 d5!





ite arrives at the estate of pure desperatio, a dagerous ream for both him ad his oppoet. The dereict archipeago of ite's paws surge forward (we, ot hs queeside paws just yet!). Tartakower gratefuy sees upo  chace ad pushes the d-paw forward. After cosiderabe expediture of capita ad resources, Back ooks ike he received precious itte from his earier ivestmets. 34  g4 Now hite's kight reay s hagg. 35 Re3 Rf5 36 N c3 Rf6 37 Re6 Kg7 38 Rxf6 Kxf6 39 d6 Ke6 40 N d5 Back's w ist so simpe after 40 Ne4! Ne5 41 K Kd5 42 f4! Nd7 43 Ng3 Ke6 ! 44 b4 Nf6 45 a4 h5 46 Ne2 Kxd6 47 Nc3 Kc6. I tried to defed this as hite agast Houdini ad ost every time. But it may ot be so easy for a huma to covert Back's positio. 40   Kxd6 41 Nxb6 h5!





The seaed move . Botvik had worked out the w  the adjourmet.

: What wn? Isn't Back the one n desperate trobe?

After a, ite has two passed paws, heped forward by the kight.

 Let's do an exercise:

 (  ): White pans to psh his passers down

the board. We must act ad act

quicky. What ca Back do?

 Pay for at e! Botvk acheicay attempts to transmte a base g-pawn nto checkate on g2. Botvk writes: "White's isfortune consists n the desperate position of his kng, under morta threat of ... h4 foowed by ... g3 and ... g2 mate. Amazngy, n every variation, Back's mate arrives before White promotes and consoidates.

42 Nc4+ 42 a4 Ne5 43 Kh2 h4 44 a Kc6 45 b4 Nf3+ 46 Khl g3 mate s. 42   Kd5 43 Ne3+ Ke4 44 a4 Kd3!





Suddey the uthikabe becomes quite thikabe: Back forces checkmate. 44 . . . g3? oy draws after 45 fxg3 Kxe3 46 a Nd4 47 a6 Nb5 48 Kh2 Kf3 49 Kxh3 Na7 50 Kh4. 45 Nd5 Ke2 46 Nf4+ Kxf2 47 Nxh3+ 47 a g3 48 Nxh3+ Kfl 49 Nf4 Ne5 ad if 50 a6 Ng4 mates ext move. 47  Kfl Aterativey, 47 . . . gxh3 48 a Kg3 49 a6 Nd2 50 a7 Ne4 51 a8Q Nf2+ 52 Kgl h2+ 53 Kfl hlQ+ 54 Qxhl Nxhl 55 b4 Nf2 56 b5 Ne4 57 b6 Nd6 ad Back ws by a sige tempo  whew!

48 Nf4 g3 49 N Not 49 Nxh5? g2 mate. 49  Kf2! The taskmaster approaches White's azy kig with whip  had. Botvik squeezes every ouce of eergy from his pieces the way a thrifty perso (i.e. my mother!) makes three cups of tea from just oe teabag.

50 a5 White's hoped-for trasformative potio turs out to be empty of magic ad chemicay ert as we. The fa promotio attempt is o more tha a ceremoia gesture. 50  h4 51 Nf4 Kfl! 52 Ng2 Or 52 a6 h3 53 a7 g2+ 54 Nxg2 hxg2 mate. The defibriator padde admisters a

ot of eectricity to the kig. His body spasms ad jerks, yet his heart refuses to restart. 52   h3 53 Ne3+ Kf2 54 Ng4+ Ke2 0-1





 strange picture to observe the embodened, hungry sma cofront the arge, chaengng throgh sheer nmbers. White's kng reazes that a resce operation w not be forthcong. There is no remedy to the comng pawn ate on g2. Here Tartakower sampes the awf trth, ike the taste of your own bood after osng a fight with the arger kid two grade eves above yours. Game  MBok-PKeres Word Championship Tournament, The HageMoscow 1 Nimzo-Indian Defence 1 4 N 2 c4 e6 3 Nc3 Bb4 4 e3 0-0 5 a3 Bxc3+

: The same as n Botv's game aganst Capabanca?

 Not qite. In Game , Capa cotted to an eary ... d7-dS. In this nstance, Keres payed the more fexibe ... 0-0, which aows his pawn strcture far more eeway.

6 bxc3

6 000 Re8!?

: What is the pont of this move?

 It's a itte odd bt actay ogica. Back pns to set p his centra pawns on dark sqares.

: What wod be the modern treatment of the ne?

 A recent game went 6 ... cS 7 Bd3 Nc6 8 Ne2 b6  e4 Ne8 10 e5 Ba6 11 Qa4 Qc8 12 Bf4 f5 13 exf6 Nxf6 14 0-0 (or smary 14 Bd6 f7 15 dxc5 Na5) 14 . . . Na5 15 B d6 f7 16 dxc5 Ne8 and Back had exceent pay, A.Ipatov-G.Kamsky, Istanb Oympiad 2012.

7 Ne2 As i hs game agast Capa, Botvk's kight heads for its optima post o g3. 7 000 eS 8 Ng3 d6 9 Be2 Nbd7?! is ooks accurate ad fais to appy sufficiet pressure upo White's cetre. Keres shoud reserve the kight for c6, with the pa: ... c7-c, ... Nc6, ... b7-b6, ... Na, . . . Ba6 ad . . . Rc8, whe Back ofte picks off the c4-paw, White's weakest k. 10 0-0 cS 11 f3! Botvik dspays deep uderstadg of the structure. He ows the bshop par ad therefore strives to reta cetra fuidity. 11 000 cxd4?!

: What is wrong with Back's ast move?

 Keres fas victim to his ove of open games n a position he shod keep cosed. He vioates the prncipe: Avoid opening the game when the opponent owns the bish pair.

Botvik writes: "Hardy a usefu decisio  the give positio as White's quees bshop ow comes to ife, ad the doubed paw s dissoved away.

Botvik was a master of ferretig out a oppoet's mistake ad recurrg patters. He oted: "Keres eary aways exchages these paws  the Nimzo Idia ... but here he shoud have refraed from dog so. Keres' move may be a s, abeit of the veia variety, ot yet a morta s  though he certay moves  that directio. 12 cxd4 Nb6 13 Bb2 exd4 14 e4!





The cotrovertibe evidece of Back's distress is strew about him. 1. White camps dow o Back's . . . d6-d5 paw break. 2. I dog so he fixes d6 as a chroic weakess. 3. White's dark-squared bishop reigs uopposed o its diagoa ad stares meacgy at f6 ad g7. 4. I compariso, White's weakeed c4-paw is easy to protect. 14  Be6 1S Rcl Re7 16 Qxd4 The quee vets her dspeasure at d6, f6 ad g7. 16  Qc7?! Botvik camoufages hs true tet (mate!) with success ad Keres, bd to the actua meace, grossy uderestimates the dager to his kg. 16 . . . Na4 shoud have bee payed. 17 cS!

: Didn't White's ast move jst dissove Back's ony pawn weakness?

 It did, bt White received a ot i return with the mutipurpose move. The cumuative end pro dct of Back's anxieties: 1. White opes the game further.

2. His cl-rook eters the game ad may swg over to g5, further stressg out Back's worried kg. 3. White dssoves hs ow paw weakess o c4. 17  dxcS 18 RxcS Qf4?! 18 ... Qd8 was a better try, but Botvik caims that after 19 Qe3! "[White's] threats ca hardy be repused. 19 Bl! Qb8? The back quee, sesg upeasat commotio  frot of her, fds hersef exasperated by her fruitess abours, so she bows strads of hair off her forehead ad staks off. After Back's ast move, the ti of the defece rus empty of fuds. 19 . . . Rd7 was his oy chace to proog resstace. 20 RgS!

The rook arrives  a swir of meag. Targets: f6 ad g7. Back udoubtedy had hopes of keepg his kgside ocet of the rook's cotamatg fuece. 20  Nbd7





 the surface Back's defeders may appear a peasat sight, but they're ot of much actua use  advacg the pot, ike biki-cad extras  a 1 950s' Evs movie. We fee it  our boes. A costrictive bad wraps aroud Back. A about him, he otices a deficit of the friedy ad a surpus of the hostie. He reaes he s soo to be evicted from  oce safe have, with othg to retur to but woe .

 ( ): Fnd hite's kngside breakthrogh

which pures Back's kgside

cover to a smooth paste.

 The rook waops g7 the way my wfe Nancy, n a daiy fit of piqe, viciosy swats her radio  aarm cock each mornng as it goes off at 6:00 a.m.

21 Rxg7+! After such a sac, for Back's kg there is o gog back to a pace of past ocece. 21  Kxg7 22 Nh5+ Kg6 The wakg wouded attempt to cross the mesa to safer territory. 22 ... 8 23 Bb2 Qe5 24 Nxf6 Qxd4+ 25 Bxd4 Kg7 26 Nxd7+ is equay hopeess for Back, whie 22 ... 8 23 Nxf6 Nxf6 24 Qxf6 forces mate. 23 Qe3! 1-0

The extet of the quees wifu evi ow grows quite apparet. Back has o remedy to the dua mate threats. Keep n mnd that Keres was a contender for the word tite n 1948! In this game (and the next!) Botvk made him appear as nythng bt a contender. Game  MBok-PKeres R Championship, Moscow 1952 Queen's Gambit Declined 1 4 N 2 c4 e6 3 Nc3 5 4 cx5

The Exchage Variatio of the Quee's Gambit Decied.

: Doesn't reease of the centra tension hep Back?

 We, I th sch decisions are a matter of stye. Personay, I aways pay the exchange nes verss the av and QGD, where I fee eqaity is't so easy for Back. Botvk writes: "This move cases Back the greatest probems, snce now a that he can contempate is a tenacios defence, whereas White has an endrng tiative.  Kasparov aso ked this ne and pied p an astronoica score with it. In essence, the ne n which yo score best is the best ne!

4 000 exd5 5 Bg5 Be7 6 e3 0-0 7 Bd3 Nbd7 8 Qc2 Re8

9 Nge2 : I see a pattern where BotvU neary aways

develops hs g-kight to the e2-

square rather than f3. Why?

 Once agan, a persona preference. BotvU deepy understood the nances of these positions which arose from the Nmzo-Indian and QGD and sed the Ng3, foowed by f2-f3 and e3-e4, pan to deady effect n his games. 9 Nf3 is an aternative set-p, and then 9 ... Nf8 10 h3 Be6 11 0-0 c6.

: What is White's pan n this position?

  person's choice of pawn strcture reveas a ot abot his personaity. Here White has a choice

Plan A: Stonewall set-up: 12 Ne5 N6d7 13 Bxe7 Rxe7 14 f4, when Wite plays Rae and then goes all out with g2-g4 next, V. Krak-O. Renet, European Cup, Clichy 1995. Plan B: Mnority attack on the queenside: 12 a3 Rc8 13 b4 N6d7 14 Bf4 Ng6 15 Bg3 Ndf8 16 Na4 Bd6 17 Bxd6 Qxd6 18 Nc5 Rc7 19 Rac Bc8 20 Rfe Qf6 21 Qd Ne6 22 Nxe6 Bxe6 23 b5 as n V.Topalov-A.Yusupov, Frakfurt (rapid) 2000. 9  Nf8 10 0-0 At the time it was an almost uheard-of plan to castle kngside. of two pans:

: What ese wod White do?

 In the 1950s, n sch set-ps, White wod caste ong and then go for an attack. In this case, Botvk repicated the same pan he sed n his game aganst Capabanca from this chapter: Ng3, then pay for f2-f3 and e3-e4. This took Keres by surprise and he faed to react propery.

10  c6

11 Rab1 Botvik toys with the mnority attack but later changes his mnd due to an naccurate plan on Keres' part. I prefer the more direct path, as n G.Kasparov-

D.Barua, Inteet (rapid) 2000: 11 f3 Ng6 12 Rad h6 13 Bxf6 Bxf6 14 Bxg6 (Kasparov is willng to hand over both bishops for Black's kights to engineer the e4-break) 14 . . . fxg6 15 e4 g5 16 e5 Be7 17 f4! gxf4 18 Nxf4 Rf8 19 Ng6 Rxf + 20 Rxf Be6 21 Ne2 Qd7 22 h4! Re8 23 Ng3 Bf7 24 Nxe7+ Rxe7 25 Nf5 Re6 26 Nd6, when White's kight dominates hs f7-bshop counterpart. 11  Bd6? It transpires ths bishop s the agent of Black's downfall. Keres, buoyed by false hope, rather navely attempts a cheapo with all the markngs of a sleazy transaction. The bishop fails to ntimidate and Keres falls prey to the temptation of . . . Bxh2+ and makes a poor decision, movng his bishop twice. 12 Khl! Ng6 12 . . . Bxh2?? fai miserably to the zwschenzug 13 Bxf6. 13 f3! The opposng factions strive and scheme n discord. Botvik alertly revises his plans and vetoes the mnority attack n favour of an central pawn break. Botvik writes: "Black can no longer prevent e3-e4, snce n the given position it is hard for him to counter with ... c6-c5./ All because of Black's lame 11 ... Bd6!. 13  Be7 Back agan. The earlier plantng on d6 produced precious little yield, so Black essentially expended two tempi for absolutely nothing n return.

: Why not pt the qestion to White's bishop with 13 ... h6  ?

 It isn't mch of a qestion snce the bishop aready knows the answer. White takes over the nitiative and attack after 14 Bxf6 Qxf6 15 e4.

14 Rbel Nd7 15 Bxe7 Rxe7 16 Ng3 Nf6 17 Qf2!





Renforcng d4, to prepare hs future e3-e4 break. 17  Be6 18 Nf5! More accurate than the immediate 18 e4 dxe4 19 fxe4 Rd7, when Black achieves some central counterplay. 18  Bxf5 Criticized by Botvik  but after 18 . . . Re8 he ntended 19 g4, when any capture on f5 would be met with g4xf5, pryng open the g-file. 19 Bxf5 Qb6 20 e4 dxe4!?

: Why did Keres aow Botvk to open the f-fie?

 This is one of those daned-f-yo-do-and-daned-f-yo-don't moments. Keres' ast move foows the prncipe: Counter in the centre when attacked on the wing. However, foowng prncipe this tme may not have been best, snce the extenatng circ tances of White's open f-fe ay take precedence. On the other hand, 20 . . . Nf8 21 e5 N6d7 22 f4 doesn't ook ke a whoe ot of fun for Back either.

21 fxe4 Rd8 22 e5! Excellent strategic judgment. Botvik gladly hands over d5 to allow  kight access on d6. I marvel at the way Botvik skiully accrues and merges multiple strategic and tactical themes; the cumulative effect leaves Black hopelessly awaiting hs fate. 22  Nd5 22 . . . Ne8 retans some pressure on d4, but in the end fails to save Black after 23 Rd1 Nf8 24 Bc2! (intending transfer to b3) 24 ... Nc7 (24 ... Ne6?? loses nstantaneously to 25 Qh4) 25 Ne4 Nd5 26 Bb3 with enduring pressure. 23 Ne4 Nf8 24 Nd6 Qc7 25 Be4!

A catalogue of Black's woes: 1. Wite clears the f-file for his major pieces. 2. White clears f5 for his kight where it eyes the sensitive g7-and h6-squares, as well as gaing a tempo on Black's e7-rook. 3. Wite may decide to plug up the d5-hole with Bxd5. 25  Ne6 : Can Back ease the pressure by sac'ng the exchange for a pawn on d6?

 Botv thoght this was Back's oy hope, bt n reaity he smpy exchanges one probem for another. Back is st bsted after 25 ... xd6 26 exd6 Qxd6 27 Bxd5! cxd5 28 xe7 Qxe7 29 Qf5 with a technicay won position.

26 Q4 Forcng weakness n Back's camp.

26  g6 Forced, but in making this move, Keres offends an entire demographic: the dark squares around hs king.

27 Bxd5 No more hole on d5. 27  cxd5 28 Rcl Chasing Black's queen into further passivity. 28  Qd7 The ageing quees dsposition deteriorates n sync with her fadng looks. 29 Rc3! Preparng transference to the kngside. 29  Rf8

 ( ): Back's position has been redced to groveng

passivity. Now

comes the time to strike. How would you contnue hite's attack?

 Add the fna attacker. Back's kng had hoped to be spared the nsfferabe anoynce of the knight's presence, bt sch was not to be his fate. Now White's heroic knight gavanzes his corades nto action.

30 Nf5! Rfe8 The kight is immune as 30 . . . gxf5?? 31 Rg3+ Ng7 32 Qf6 mates next move. If nstead, 30 . . . Ree8 31 Nh6+ Kh8 (or 31 . . . Kg7 32 Qf6+ ! Kxh6 33 Rh3 mate) 32 Qf6+ Ng7 33 Nxf7+, Black must fork over the exchange, snce 33 ... Kg8 34 Nh6+ Kh8 35 Qxf8+ Rxf8 36 Rxf8 is mate agan. 31 Nh6+! The kight thunders past. More cold-bloodedly effective than merely takng the exchange  h6 is the puncture wound through which Black's misery drips. ith the feast yet to come, Botvik is dsnclned to be bought off with gifts of scraps from Black's table. 31  Kf8 31 . . . Kg7 is once agan met by 32 Qf6+ Kxh6 (the kng wanders off to who-knows where, like a child who plays hooky from school) 33 Rh3 mate.

32 Qf6 Threateng mate on the move. 32  Ng7

Black's overcrowded pieces scrunch in close.

33 Rcf3

Game over. White threatens mate, begng with Qxf7+!. In this hopelessly busted position, even Black's most sincere efforts are rendered moot. 33   Rc 34 Nxf7 Re6 35 Qg5 Nf5 36 Nh6 Qg7 37 g4 1-0 Back's kng crnges, beeagered by the swir of hostes a around him, and he drops a piece as we, amost as an afterthoght. Game  MBok-LPosch Monte Caro 16 English Opening

At the time of this game, Portisch was a contender for the Word Championship, whe Botv's career entered its twiight.

1 c4 e5 2 Nc3 Nf6 3 g3 d5 4 cxd5 Nxd5 5 Bg2





The Reversed Dragon. Botvik once wrote that the Open Sicilian position was too potent an openg to allow Wite to attan it a full move up.

: Is this tre?

 obaby not. Top GMs contne to aow the Reversed Dragon to this day and score jst fne with the back side. However, Back mst pay one of the qiet, po sitioa nes. If he goes for some tra sharp Ygosav Attack variation, the extra tempo trns deady potent aganst him.

S 000 Be6 GMs play 5 ... Nb6 today.

: What is the dfference?

 It is n Back's best nterest to keep a firm grip over the d4-sqare nd not aow White to pay the move n one go.

6 Nf3 Nc6 7 0-0 Nb6 The most accurate move. hite gets a clear edge  allowed d2-d4. For example: 7 ... Be7 8 d4! Nxc3 9 bxc3 e4 (or 9 ... exd4 10 Nxd4 Nxd4 11 cxd4 c6 12 Rb1 Qd7 13 Qc2 with advantage to hite, who dominates the centre and may apply pressure down the open b-and c-files, L.Chrstiansen-E.Handoko, Surakarta 1982) 10 Nd2 f5 11 e3 0-0 12 c4 and hite's extra central control gave him the edge, G.Kasparov V.Korchnoi, Paris (rapid) 1990.

8 d3 Be7 9 a3

9 000 as?! It sn't worth weakeng the queenside to prevent b2-b4. Current theory has Black playng ... a7-a5 oly in response to hite's b2-b4. For example: 9 ... 0-0 10 b4 f6 11 Bb2 a (the correct timng) 12 b5 Nd4 13 Nd2 c6 14 bxc6 Nxc6 15 Nb5 a4 16 Rc1 Ra5 with mutual chances, E.Bacrot-V.Topalov, Dubai (rapid) 2002.

10 Be3 0-0 11 Na4 A standard Dragon manoeuvre. The swap of the b6-kight helps hite out n his queenside ambitions.

: Doesn't takng on b6 daage Back's pawn strctre?

  itte, bt Back is compensated n the form of ncreased dark sqare contro and the bishop pair; eg 11 Bxb6 cxb6 12 Rc1 f6 13 Nd2 Rc8 14 Nc4 7, when I prefer White bt Back's game is qite payabe as we, Zhao Xe-B.Ydz, FIDE Grand Prx, Nanjng 2009.

11  Nxa4 G.Kasparov-Kr.Georgiev, orld Blitz Championship, Saint Jo 1988, saw 11 Nd5 12 Bc5 Bd6 13 Rc h6 14 Nd2 Rc8 15 Ne4 and, just as hite began to exert a degree of queenside pressure, Black blundered with 15 ... b6? which allowed Kasparov a combnation.

 ( ): White to pay and wn ateria.

 Defection. 16 Nxd6! cxd6 17 Bxb6! Nxb6 18 xc6, pickng off a pawn.

12 Qxa4 Bd5 13 Rfc1 : Why not the a- rook?

 Bo tv's attack ( for now) is on the qeenside, so he ogicay tiizes both of his rook s n that sector of the board .

13  Re8 14 Rc2 Bf8 Labelled "complacency by Botvik, who gave 14 ... Bd6 15 Nd2 Bxg2 16 Kxg2 as Black's best chance to defend his queenside.

15 Rac1 Now hite may contemplate Nd2 and, after a bshop swap, a potentially damagng exchange sac on c6. 15  N8

·



elcome, welcome! Please enter my humble home. If we take on c7, then Portsch plans to slam the door in our face with ... Bc6, trapping our rook. If we dot take, then Black plans ... c7-c6, rendering our doubled rooks rather silly on the c-file.

 ( ): Should we accept Black's dare and enter with 16 Rxc7  ?

 Black's last move fails to fit the position's requrements. We absolutely should take the dare. Black's last move was a blnder. Rare is the occasion when one side is able to retreat a fully developed piece back to its home square i an open position and get away with the crme. His knight retreat proves to be a dangeros cocktal, high on ambition and short on effectiveness.

16 Rxc7! Apparently both parties are amenable to the deal. 16   Bc6 17 R1xc6 bxc6

 ( ): White can play 1 Rb7, when he

gets excellent compensation for

the exchange. Instead, he found an nfitely stronger contnuation. hat was it?

 The herent weakness of f7 wafts its npleasant odor, ntil it permeates Black's camp. How often do you see a game where one side gets the good fortne to offer both rooks?

18 Rxf7!! This move must have come as a jarrng shock to Portsch. Now we sense a radical shift of energy, Black's decreasng and White's ncreasng with each move. 18  h6 The rook's firm grip is met by the black kng's dreaded limp-fish handshake. Declng is the oly way to contnue playng.

: Why didn't Black accept?

 Acceptance leads to slaughter fter 18 ... Kxf7? (the old kng, overestatng his grip on power, ndulges n a poor decision) 19 Qc4+ Kg6 (other moves fail to come nto consideration 19 ... Ke7? 20 Bg5+, 19 ... Kf6? 20 Bg5+, or 19 ... Qd5? 20 Ng5+) 20 Qe4+ Kf7 21 Ng5+ Ke7 22 Qxe5+ Kd7 23 Bh3+ and Black's plate is so empty that he sees his own reflection.

19 Rb7 Qc8 20 Qc4+ Kh8 20 . . . Qe6 21 Nxe5 Qxc4 22 Nxc4 is ao completely lost for Black. White simply gets too many pawns for the exchange. 21 Nh4!





Yet another punshng foray  g6 is punctured. Now Botvik's pieces spit forth. 21  Qxb7 The queen strives to expunge the unpleasant association with Wite's ntrudng rook from her memory. 22 Ng6+ Kh7 23 Be4 Bd6 24 Nxe5+ g6 24 . . . 8 (the hauntng images dancng within the black kng's mnd feel more real than actual reality) 25 Nf7+ wns the queen. 25 Bxg6+ Kg7

 ( ): The once rich black kng's frnishngs are now as

Spartan as a mok's

in a monastery. How did Botvk end all resistance?

 The crushng tide of events press ncomfortably upon Black's kng, who is cast out of his doman to wander n dngerous places.

26 Bxh6+! 1-0 After 26 . . . Kxh6 27 Qh4+ Kg7 28 Qh7+ Kf8 (the kng falls to his knees and prays to some sant who specialzes n rescung believers from lost causes) 29 Qxb7 (the Botvik giveth and the Botvik taketh away; the duration of the black queen's life comes to a tragic and sudden end, as she slumps forward) 29 ... Re7 30 Qxa8 does the job. I was about to say somethng clever here but fnd that I am unable.

: Don't tell me. Let me guess. Could it be that for the thrd time n just

are rendered mute, bedazzled by Botvk's attackng prowess?

 ...

one chapter you

Chapter Two Botvinnik on Defence

I had the black pieces n this position from a 1977 simultaneous game aganst Botvik. Fortune fnally smiled on me. Just look at Wite's trapped g-bshop and rook, glued to ther posts ndefitely, and his pathetic, stray, sick lamb of an h pawn, about to fall by the wayside. If victorious, I imagned a wondrous destny: 1. The city of Montreal throws a parade n my honour. 2. Popular kids at high school remorsefully seek absolution for past ndiscretions toward your humble writer, and high five me ncessantly. 3. Curvy high school cheerleaders, aghast at not earlier recognng my unseen depths, fall tearfully to their knees, hands prayerfully clasped, implorng forgiveness and beggng me for dates. 4. My stubborn case of acne fnally clears up. In the game, Botvk's position stretched elastically, yet never snapped. Alas, none of my dreams came to pass  wait; my mstake; the acne did ndeed clear up  snce Botvik hung on like grim death, complicated nto a frestorm, and n tacit parership with your writer's ncompetence, managed rudely to steal the pont from your deservng writer. When you are so utterly outmatched (even with receivng simul odds) by a legendary opponent, it's not such a bad idea to factor n crushng disappontment. Nevertheless, I remember leavng the buildng teary eyed. I discovered, as did Botv's opponents n this chapter, that achievng a superior or wng position agast Botv didn't necessarily equate to actually scorng the pont aganst the legend. In this chapter, we examne just how Botvk regans his mojo when nder attack or struggng n an nferior position. Game 4 AKotov-MBotvk  Championship, Moscow 1944 Queen's Indian Defence

Who among us has not read Kotov's classic boo k, Think Like a Grandmater   1 4 N 2 Nf3 b6 3 e3 c5 4 B3 Bb7 5 c4 e6 6 0-0 Be7 7 Nc3 7 b3 leads to Zukertort Colle positions.

7  5!? Perhaps this move is playable, but I don't lke it. : Why not? Black simply chalenges the centre.  He shold't be cofrontng White and openg the game when laggng n development. Instead: a) The very natral 7 ... O-O?! ay be a strategic error as well. White forces an advntage after 8 d5! exd5 9 cxd5 d6 (or 9 . . . Nxd5 10 Nxd5 Bxd5 11 Bxh7+ Kxh7 12 Qxd5 and White's grip on d5 nets a clear advantage) 10 e4 a6 11 a4 with a good Benoi, snce Black's ight-squared bishop hits a pawn wal on d5, V.Malaik-M.Smantsev, Mnsk 1998. b) 7 . . . cxd4 8 exd4 d5 9 cxd5 Nxd5 10 Ne 5! 0-0 11 Qh5 g6 12 Qh3 looks ike a dangerous version of an isolai for Black, E.Daielian-M.Brodsky, Cappelle la Grande 2006. c) 7 ... h6! : What is the pont of Black's last move?  I lke this idea, which deies White d4-d5 tricks by removng the Bxh7+ theme. After 8 b3 0-0 9 Bb2 cxd4, Black is okay whichever way White recaptres, R.Edouard-Al.David, Le Port Marly (rapid) 2012. Pnciple Open the game when leading in deelopment.

8 cxd5 exd5 Botv always favored central control over hypermodern piece play. 8 ... Nxd5 9 Nxd5 isn't pleasant for Black either. J.Akesson-K.Plgaard, Gothenbrg 2003, contnued 9 ... Qxd5 10 e4 Qd8 11 dxc5 Bxc5, when 12 b4! looks tough for Black. 9 Bb5+! Pnciple Conont the opponent when leing in deelopment. 9  Kf8? If the position were given n EEG, the branwaves produced wold be reliably abnormal. This overreaction looks ike a case of the cre exceedng the discofort of the orignal alment. 9 ... Nbd7 10 Ne5, npleasant as it appears, may be superior to sef-denyng castng. 10 b3!

Dual prpose: 1. Kotov prepares to fiancheUo his dark-squared bishop. 2. This, n trn, discorages Black from the ravelng plan ... g7-g6?! and ... Kg7, snce it would put his kng n the drect fre of White's fincheUoed bishop. Botv's openg has been an nquestioned disaster! 10  a6 11 Be2 Nc6 12 Bb2 Rc8 13 Ne5 I don't th this is White's most drect plan.

: What would you suggest?

 With Black's kng on f8, his h8-rook wll be out of play for quite some time. White's best bet is to cofront n the centre and on the queenside as soon as possible, when Black is essentially short five fightng ts, due to his AWL h8-rook. Here 13 Rc1 h5 14 dxc5! looks exceedngly dfficlt for Black, whose position, n the throes of dissolution on every front, hangs by a worn thread. 13  B6 14 Nxc6 Rxc6 15 Bf3!? White retans a clear advantage after the correct 15 dxc5!. The text looks ike a case of too much preparation. Kotov's move, a by-pro duct of the an idea  lke foa's relationship to beer n the mug  may be a sub-optmal choice n a very proisng position.

 (  ): How did Botv anage to reieve a good ch of his troubles with his next move?

 Pnciple Close the position when lagging in deelopment. 15   4! 16 g3 Now White threatens to chop on d5. 16   R8 17 bx4 Rx4 18 Q 3! Q8 Botv, stil well short of equality, covers aganst cheapos such as 18 ... g6?? 19 Nxd5!, pickng off an important central pawn. The alternative was 18 ... Bb4, chalengng an enabler of White's e3-e4! break; but even then White retans the advantage by openg the position once agan with 19 Nxd5! Bxd5 20 e4 Be6 21 e5 Ne8 22 d5 b5 23 fc1 Rxc1 + 24 xc1 which gives hm a promisng attack, despite Black's extra piece. 19 Rac1?! /IAn idle nd is the deviY s playgrond !/ so my thrd grade ath teacher wold admoish me when she caught me daydreamng of a athless world, whle gazng out the wndow at the enticng playgrond. ne can oly amend and cross-reference a pln so much and then the need for action overrides the need to edit. No need to cook the banana. Just peel nd eat. But Kotov vacilates. The e4-square is the governg body of the position, where authority flows down. White's goal gets bogged down n a series of ntermnable delays and dfficties  al necessary and al sef-ficted. The tme is now to seze the itiative with 19 e4! Nxe4 (Black also reans on the defensive n the lne 19 ... xc3 20 Qxc3 Qxc3 21 Bxc3 dxe4 22 Bg2 Ke7 23 l Bc7 24 Ba5! Nd7 25 Bh3 Rb8 26 Bxd7 bxa5 27 Ba4) 20 Nxe4 dxe4 21 Bxe4 Bxe4 22 Qxe4 Qc6 23 Qf5 Qc8 24 Qf3, when the sllen h8-rook wll probably cost Black the game. 19  Qe6! Fightng for e4, whle retang control over c4. 20 Bg2 h5!

Intendng ... h5-h4, a move which accrately reflects the ncreasngly hostile tnge to Botvk ambitions. He attempts to brazen it out by moldng a virtue from a vice, threateng to activate the h8rook nd lanch an nexpected kngside attack.

21 Ne2 Now an e3-e4 break wil be near mpossible to achieve. I wold have tossed n 21 h4. 21  b5 And here perhaps the imediate 21 ... h4 shold be played.

22 Nf4 Qe7 23 Qd1 The last chance for 23 h4. 23  h4 The once-heavy volme of White threats decnes to zero. Kotov's sluggish iddlegame play allowed Botvk to ake fl repars and suddely I prefer Black's position. 24 Qf3 Kg8 25 Rf1 N e4 26 N 3 Rh6!

The sleepng rook awakens. 27 Qe2 hxg3 28 fxg3?! The wrong recaptre. White volntarly accepts a weak, backward e-pawn. 28  Qg5 29 Bxe4? This move is ndisputable evidence that White's advantage, now long gone, took a wrong trn. Kotov removes the offendng knight, but n so dong, fatally weakens his ight squares. 29   xe4 30 Nf4

 (  ): Just compare Black's position to his dismal one after the openng! What a change. Come up with a plan to break White's resistance.

 Principle Opposite-coloured bishops favour the attacr. 30  Bxf4! The bishop greets his wayward brother with precious little warmth or affection. 31 exf4 Q5! Dual prpose: 1. Blockadng White's d-pawn. 2. Threateng to eash ight square pan with . . . e4-e3 next. 32 Qg2 Rhc6 33 Rxc4 Rxc4 34 h3 b4! Advancng his majority and fixng a2 as a target as well. 35 Kh2?

·



White is at least consoled with an ilusion of safety. His kng position, now absent of defenders, stands dereict.

 ( ): In fact Kotov just blndered n a busted position. How did Botv exploit it?

 The start of a zwischezug combation. Botv concludes crisply, without wasted motion. 35  e3! 36 Qx5 Rc2+! There it is: zwischezug! Black's rook sneaks n and puts fnger to ips n a  slence" gestre to White's forces. He takes control over the seventh rak, renderng White helpless.

37 Kgl The nbeievng kng, who up ntil now had oly seen the nside of chrches and temples via fnerals and weddngs, decides to engage upon spiritual matters. 37   Bx5 0-1 White cn't save his bishop, snce 38 B is met by 38 ... Rg2+ 39 Kf1 (the cofused kng's eyes glaze over, ike discont chnaware) 39 ... e2+. Game 5 DBroste-MBotvk World Championship (18th atchgame), Moscow 1951 emi-lav Defence 1 4 ne of the great shortcongs of lfe is that most people rarely achieve their heart's desre, ess blnd luck plays a role. Had Bronsten won this game, he would very ikely have defeated Botvk n their world championship match and had his name enshrned n history, next to the lkes of Morphy, Lasker, Capablanca, Botvk, Fischer, Karpov and Kasparov. fortnately, Bronsten's fe fell a siver short of his dreams. 1  5 2 c4 c6 3 Nc3 N 4 Nf3 e6 5 e3 a6

·



This lne contnues to be played by GMs, but it's a thakless task for Black, n my opon, and I believe 6 ... Nbd7, the an lne, is clearly a superior choice. I was tempted to give Black's move a ?!" ark but reaized it would outrage al the people who play this ne, so I chickened out!

: Can you explan this move?

 Black cn either play for an mmediate ... b7-b5, or he can meet 6 Bd3 with 6 ... dxc4 7 Bxc4 b5 8 Bb3 b5, transposng to a Queen's Gambit Accepted. In my opion, the . . . a7-a6 nes wor k better n a pure Slav rather than n a Semi-Slav where Black is already comitted to ... e7-e6. Black's light-squared bishop normally belongs on the outside of the pawn chan.

6 Bd3 The mmediate 6 b3 is White's best shot at an advntage. G. Kasparov-B. Gefand, Lnares 1991, contnued 6 . . . Bb4 7 Bd2 Nbd7 8 Bd3 0-0 9 0-0 Bd6 10 e4 dxc4 11 bxc4 e5 12 c! Bc7 13 Na4 exd4 14 h3 Re8 15 Re h6 16 Rb Nh7 17 Bc4 Qf6 18 Rb3 Nhf8 19 B Ng6 20 Nxd4 (f3 is n the air) 20 . . . Nde5 21 Bf Rd8 22 Bb2 and I prefer White, whose kngside majority feels more potent and his pieces look more threateng than Black's. 6  b5 6 . . . dxc4 7 Bxc4 b5 followed by . . . c6-c5, transposes to Queen's Gambit Accepted. But Botvk probably decned to enter for two reasons: 1. Bronsten was a monster n open positions. 2. Botvk himsef lked to take on the isolated d-pawn positions, with ther resultng kngside attackng chances. So, due to this bias, he may have been somewhat hesitant to enter from the Black side. 7 b3! A good reaction. White may recapture on c4 with his b-pawn. 7  Nb7 8 0-0 Bb7

9 c! White's extra space gives him a sight but endrng advantage. Bronsten's move is stronger than retanng the central tesion with somethng lke 9 Bb2 Bd6 (now Black's bishop reaches a superior square) 10 Qe2 (it isn't too late for 10 c!) 10 ... bxc4! 11 bxc4 dxc4! 12 Bxc4 c 13 fd cxd4 14 xd4 Bc5 15 Rd2 Qe7 16 Rad 0-0 and Black equaized, D.Zlbersten-V.Akobian, S Championship, Sn Diego 2004. 9  Be7 : Can Black free his po sition with an . . . e6-e5 trick?

 This violates the prnciple: Avoid conontation when behind in development. 9 ... e5? 10 Nxe5 Nxe5 11 dxe5 Nd7 and now 12 e6! looks good for White after 12 ... Nxc5 (12 ... fxe6? 13 Qh5+ Ke7 14 e4 is obviously awful for Black) 13 exf7+ Kxf7 14 Bc2, when Black reans behnd n development n an open position. 10 a3 a 11 Bb2 0-0 12 Qc2 g6 : Is this weakeng move necessary?

 Sooner or later it wll be. Black wants to retan options of ... Nxe5 n case of Ne5 from White. 13 b4 axb4 14 axb4 Qc7?! Black should jump on every po ssible exchange, startng with 14 . . . xa 15 xa Qc7, folowed by . . . Ra8. 15 Rae1!





In the presence of my mother, I use the word shoot!" when I'm anoyed; when at home, yor foulmouthed writer often uses a siilar sondng but dfferent word. My feeng is that Botv's thoughts

were clearly not along the lnes of  shoot" about here! Bronsten deftly retans rooks. Principle The side with extra space should avoid swaps. Botv is gettng squeezed: 1. The . . . e7-e5 break, his oly sorce of conterplay, remans an impossible dream. 2. A closed centre means that White gets to attack with ease with ideas ike Ne5 and f2-f4. 3. Black's light-squared bishop is a disgrace to the brotherhood of bishops, as he contnues to sit n Zen lke silence on b7.

: Well, i this is the case, then just where did Botvk go wrong?

 Honestly, I thnk his istake was enterng this lemon of a lne n the frst place. 15   Rfe 16 Ne2 Bf 17 h3 Bg7 1 Ne5 Nf Botv decides upon passive defence, snce cofrontation favors his opponent after 18 ... Nxe5 19 dxe5 Nd7 2 0 f4 f6 21 exf6 Nxf6 (Black must recaptre with his kight snce 21 . . . Bxf6?? gets slaughtered by 22 Bxg6) 22 Be5 Qd8 23 Nd4 Nd7 24 Bxg7 Kxg7 25 e4! e5 26 Nf3 d4 27 Qd2 Bc8 28 Rd and Black wil eventually drop d4.

19 £3 N6d7 20 £4 £6 The imediate 20 ... f5. ntendng ... Nf6-e4, is a thought.

21 N£3 Re7 22 Nc3 £5 Achievng ... f7-(f6)-f5 has not brought about equaity. 23 Ra1!

: Didn't you say earlier that the side with space shouldn't swap?

 Flexiblity is important. Sometimes you must deiberately violate a prnciple i you believe the situation is n exception. By achievng a sei-locked kngside, Black defended well aganst a cong attack on that flak, so Bronsten switches gears and trns his attention to the queenside. 23   Ree 24 Ne5 Rxa1 25 Rxa1 Ra 26 Qb1!

: Why an exclam?

 The queen frees the b2-bishop for duty elsewhere. I can't say nymore about the move or I risk givng away the answer to the comng exercise! 26  Qc?!

·



Botv, with his last move, akes a rare strategic evaluation iudgment. He should have played his queen to b8 nstead. Cong up with a viable offesive pln n such a situation is no smple task. Hypotheses rapidly form and as quickly die, klled off by ther own future implications. Bronsten enters a hidden portal of thought, discoverng a solution.

 ( ): How would you play for the wn as White n this position?

 Give up a piece for two conected passed pawns. 27 Bxb5!!

: Isn't White n reality makng purchases he can't afford?

 Bronsten's move shows brlint judgment. White's sac sets n motion a new chan of thought. Kasparov writes: /IA purely positional sacrfice: the par of conected b-and c-pawns wll be stronger thn the bishop. White's dynaic evaluation proves more correct./ White is probably wng at this stage, but as n horse racng, there is no such thng as a sure wner. 27   Nxe5 28 e5 Bh6 29 B! Now we see the hidden pont of 26 Qb! . 29  cxb5 30 Nxb5 N7 31 N6 And now we see why Botv's 26th move was naccurate. White's kight, who smears d6 with his very presence so deep n Black's territory, arrives with tempo. 31  Rxa1 32 Qxa1 Qa8

33 Qc3 Botv felt that 33 Qb2 was more accurate and led to a wn, while Kasparov vigorously disputed the claim, givng lengthy analysis on how Black should be able to hold.

: And your opon on the atter?

 As I plead to my mother nd sister, when they erupt nto an argment and both try nd wo o me to ther camp: Keep me out of it!" I'm neutral, ike the Swiss! Actualy, the writer's declaration of the dreaded  nclear11 is n a sense without pertnence to the actual truth over the board and, nstead, may simply be a testmoial to the iitations of your writer's not-so-expansive mnd to comprehend. 33  Bf8 34 b5 Bx6 No choice n the atter. Now Bronsten gets three deeply entrenched passers.

35 exd6 Qa4 36 Qb2 : hy not 36 c6 - ?

 Actually, I was about to lectre you and tel you that yor suggestion would be an answer to Botvk's prayers, snce Black angles to sac his piece back to einate White's pawn arada. However, lookng deeper nto the lne, apparently yor suggestion ay be a double exclam, which ay wn for White! I realze that most readers don't go over detaled analysis, but I rge you to do so here. I learned a lot studyng the br eakthrough technique. The analysis: 36 ... Qxb5 37 cxb7! (37 cxd7? Qxd7 is a near-certan draw) 37 ... Qxb7 38 Qc7 Qxc7 39 dxc7 Nb6 40 Kf2 Kf7 41 Kf3 g5 (Black must keep White's kng out of the kngside, but apparently he canot do so forever) 42 Ba3 Ke8 43 Bc5! (tang the would-be rescuer kight, who offends more than soothes the harried defenders) 43 . . . Nc8 44 h4! (White must fnd a way to pnctre the kngside) 44 . . . h6 (44 .. . gxh4? 45 Kf4 Kd7 46 Ke5 is an easy wn for White) 45 h5!





(a key move: White fxes h6 as an eternal target; this mens Black's kng must reman on the kngside) 45 . . . Kf7 46 Ke2! (now White's kng fltrates on the other wng) 46 . . . Ke8 47 Kd3 g4 48 Kc3 Kd7 (otherwise White's kng simply arches to b7) 49 Bf8 Kxc7 50 Bxh6 Kd7 51 Bf4 Ke7 52 K4 Nb6 53 h6 Kf6 54 Kc5 Nc4 55 Kc6 Kf7.

 (  ): How does White wn?

 Deflection. Black's kng feels constriction of the chest and a queasy stomach at the thought of the nvaders at his gate. After 56 h7! Kg7 57 Kd7, B lack's pawns fall. 36  K 37 Kh2 Bronsten was critical of his own move and suggested the plan 37 Bd2 and Qb4. And please don't ask me what I thk. I just appied for Swiss citzenship. 37  h6 38 e4!





Pawns mesh nd clot n the central tngle. Bronsten does his con artist best to contrive deceptive pitfals for his opponent. The best practical chance", accordng to Kasparov. 38  f4? Here, Botv claimed that 38 ... dxe4 39 d5 ws for White, but Kasparov correctly disputes the evaluation, citng 39 ... e3!! (this is not a far fight, snce Kasparov gets to use computers while poor Botvk is on his own for the analysis). After 40 c6 (Hdini suggestion) 40 ... Qf4+ 41 Kg Bc8 42 dxe6+ Kxe6 43 Qb3+ Kxd6 44 Bxe3 Qe4 45 Bf2 Nc5 46 Qg8 Qb + 47 2 Qxb5 48 Bg3+ Kxc6 49 Qxc8+ K6 Black stll struggles but may yet hold the game.

39 e5 g5 40 Qe2 Kg7 White wns after 40 . . . Qxd4. For example: 41 c6 Qxe5 42 Qxe5 Nxe5 43 cxb7 Nd7  Bd2! e5 45 Ba5 d4 46 Kg e4 47 Bc7 d3 48 Kf h5 49 b8Q Nxb8 50 Bxb8 Ke6 51 b6 Kd7 (or 51 ... g4 52 d7 Kxd7 53 Bxf4) 52 Ke h4 53 b7! Kc6 54 d7 Kxd7 55 Bxf4 etc.





In this critical position the game was adjorned and no one, mysef ncluded, had any doubts that White would wn," wrote Botvk. 41 Q3? Bronsten sealed this move, which raises more questions than answers, nd allowed Botvk a mraculous escape. 41 h4! is completely wng for White after 41 ... Qxd4 42 c6 f3 43 gxf3 Qxh4+ 44 Kg2 Bc8 45 cxd7 Bxd7 46 b6 Bc6 47 d7. 41  Nb8!! The queenside is ittered with warrng fiefdoms and mcipalities, all deterned to estabish pre emnence over the others. Botv fnds n ncredibly hidden fortress defence. The key is control over the critical c6-square. 42 h4!

White's best chance: strip Black's kng of shelter. 42  Qc4 43 Qh3!

43  Qxb5! The pawn is more mportant than the bishop! After 43 ... Qxc1?  hxg5 hxg5 45 Qxe6 Qe3 46 Qf6+ Kh7 47 Qxg5 Qg3+ 48 Qxg3+ fxg3+ 49 Kxg3 White wns; or i 45 . . . f3 46 Qf6+ Kg8 47 e6 fxg2, then 48 Qf7+! (not yet 48 Kxg2?? Qc2+ 49 Kg3 Qd3+ 50 Kg4 Qe2+ 51 Kxg5 Qg2+ and draws) 48 ... 8 49 Kxg2 (now there is no perpetual check) 49 ... Qc2+ 50 Kg3 Qd3+ 51 Kg4 Qe2+ (or 51 ... Qxd4+ 52 Kf5 Qf4+ 53 Kg6 or 51 ... Bc8 52 Qf8+ 7 53 Qf5+) 52 Kxg5 Qg2+ 53 Kf6 and the white kng escapes. 4 hxg5 hxg5 45 Qxe6 Q3! The two camps fng blows n each other's direction with mutual vehemence. Inspired disorder is a form of energy which should never be nderestated on the chess board. Botvk is stll withn his adjorment analysis at this stage and tidies up with a bit of housekeepng. The threat is a perpetual check, startng with ... Qg3+ next. 46 Qf6+ Kh7 47 Q+ Botv's deep pont: after 47 Qxg5 Qg3+! 48 Qxg3 fxg3+ 49 Kxg3 Bc8 50 Kf4 Kg6, Black achieves a fortress draw. Compare this with the variation 43 ... Qxc1? 44 hxg5 hxg5 45 Qxe6 Qe3 etc above. Here White has n extra piece (nstead of a pawn on b5), but cn now oly draw because he canot contest the light squares. 47  Kh8 48 Qf6+ Kh7

49 Bxf4 Redoubng one's efforts nto a hopeless cause (wng!) oly adds to one's frustration. Admittedly, Bronsten's move, Wagnerian stuff, displays theatrics more than the nfiction of any real damage. But once agan, after 49 Qxg5 Qg3+! 50 Qxg3 (White's queen raises a soprano outcry over the ntrusion nto her chamber) 50 ... fxg3+ 51 Kxg3 Bc8! 52 Kf4 Kg6 53 g4 Nc6, Black mraculously achieves his fortress and a

draw. 49  gxf4 50 Qf7+ Kh8 51 Qe8+ f course 51 Qxb7 leads to imediate perpetual check by 51 . . . Qg3+ 52 Kg Qe + etc. 51   Kg7 52 Qe7+ Kh8 53 Qe8+ Kg7 54 Qe7+ The queen professes her tender, eternal love for her beloved, Black's kng. His cold response: Whatever ." 54   Kh8 55 Qf8+ Kh7 56 Q+ Kh8 5 7 Qxb7 : Can White gamble nd just take on f4?

 It's stll a draw after 57 Qxf4 Qh7+ 58 Kg Qb + 59 Kf2 Qc2+. 57  Qg3+ The iracle arises and White's kng lacks a graceful avenue of escape from his perpetual predicament.

58 Kh1 /- our bleary-eyed writer nvested more time nto this game than any other n the book. After ther precipitous clmb, the white pawns faled to reach the sut. It's scary to thnk about just how close Botvk came to losng the tied World Championship match aganst his great rival, Bronsten. Game  NPaevsky-MBotvk Alekhne Memorial, Moscow 1956 icilian Dence 1 e4 c5 2 Nf3 Nc6 3 4 cx4 4 Nx4 Nf6 5 Nc3 6 6 Bc4

·



The Sozn Siclian. White's c4-bishop development scheme was a lfetime favourite with Bobby Fischer. 6  e6 7 0-0 This tends to be a quieter lne (although certanly not n this game!).

: How can White play more aggressively?

 By enterng the psychotic Veiirovic Attack with 7 Be3 Be7 8 Qe2 a6 9 0-0-0 Qc7 10 Bb3 0-0 11 hg Nd7 12 g4 Nc, when a standard contnuation is just to g o for it with 13 Nf! ? 7   Be 7 8 Be3 0-0 9 Bb3 Here J.Ems-A.Grischuk, Esbjerg 2000, went 9  a6 10 a4 Qc7 11 Qe2 Bd7 12 f4 Rac8 13 Ba2 Nxd4 14 Bxd4 (now White threatens e4-e with attackng chances) 14 ... e 15 Be3 Be6 16 a! Bxa2 17 xa2 Qc4 18 Qf3 exf4 19 Bxf4, and n this equal position my editor John went on t o score a draw aganst his ntiidatng opponent. 9  Na5 Botv goes after the bishop pair. 10 f4 b6!? After 10 . . . a6 11 e dxe 12 fxe Nd 13 Nxd exd 14 Qf3 Be6 15 Rad Nxb3 16 axb3 White's control over d4 gave him a tny edge, J.Vn der Wiel-F.Nijboer, Dutch Championship, Rotterdam 1997.

11 Qf3 It doesn't ake much sense to me to develop the queen n the path of a b7-bishop's angry gaze. 11 e is probably White's oly hope for an edge n this position. 11   Bb7 12 g4?!

·



Gamblng usualy represents a vice ess one possesses a srplus of wealth. The g-pawn, determned to defy common sense, ateraly veers off to attack, all by himsef.

: Isn't his aggre ssive g2-g4 push stndard n Siclians?

 Aggressive doesn't always equate with effective or sond. This overly risky attempt to distrb the natral order of lfe on the kngside is misplaced, snce the white bishop normally sits on f3 n Schevenngen Siclians where White tosses n g2-g4. In this nstance, the e4-pawn lacks the necessary protection. White's entire attack appears more possessed by passion and fry, thn by exact execution. 12  R The most exact move. 12 ... d is also possible. Principle Counter in the centre when menaced on the wing. Mok Tze Meng-Mrugan, Kuala Lmpr 1996, contnued 13 e Ne4 14 f Bc, when I lke Black's chances  but then agan, he may get ated! 13 g5 Rx3!

Black's exchnge sac is both theatic and sond. Principle trike in the centre when assaulted on the wing. Here the road divides.

 ( ): White can accept the rook with 14 bxc3, or he cn enter the compications of

14 gxf6. ne way leads to near-dynaic equality; n the other he is busted. Calcate and evaluate both lnes, and make yor choice.

14 bxc3?  He should take the f6-kight rather than the rook. White is slightly worse but stll reans n the game after 14 gxf6! xe3 15 Qxe3 ( 15 fxe7? xf3 16 exd8Q xf + 17 xf xd8 leaves White a pawn down nd busted) 15 ... Bxf6, as n P.Poutianen-Z.ibli, World Jor Championships, Athens 1971. I stll prefer Black here, but this is nfitely better than what Padevsky got n the actual game. 14  Nxe4 White's troubles: 1. Black domnates the light squares and the h-a8 diagonal. His ight-squared bishop's capablities, when contrasted with any of White's mnor pieces, seem godike n comparison. The bishop goes on to become an eforcer, an strument of fate, destned to brng on cataclysic chnges upon White. 2. This n turn means White's kng is n a lot more danger than Black's. 3. White's queenside structure is ruptured and weak. 4. With a pawn for the exchange, Black isn't realy even down aterial. 15 Qg4 Qc8! 16 Rf3 Plang to load up on the h-fle and target h7.

: Didn't Black just blnder on his last move? White now can play 16 Nxe6.

 When you set a trap be careful not to be the one trapped. White's shot on e6 is a big blnder, snce Black coolly responds with 16 ... d5! nd wns the rash white kight. 16   Nxb3!? I'm not sure why Botv decided to repair White's damaged queeside. I would have avoided it and played 16 ... g6. 17 axb3 f5!





Black is ready, sittng on his rockng chair on the front porch, pump-action shotgn over his knees. 18 Qh4? Not 18 gxf6?! xf6 when it is Black, not White, who attacks. But 18 Qh3! would have prevented Black's next move. It saves us a lot of grief i we avoid a dubious idea n the frst place, rather than later try and fx a past error. The delusional queen istakely believes that by divne decree, she has been granted domon over the kngside and all its habitants. The wrong square, but White has aready savoured the drug of attack n his vens nd now deands more. 18  e5! Mutual ambitions meet and overlap. This move effectively cancels multiple hostile variables, klls them before their seed has a chance to take root. Time to take a stand. White's kngside buld-up, i left nchecked, could spel trouble i his opponent gets careless  somethng Botvk almost never did . Black contnues to distract White vigorously n the centre n anticipation of the comng wng attack on h7.

19 R3

The tny flicker of hope's flame soon dies out n the attack which never was. 19 xa7 exd4 20 Bxd4 Qc6 21 xb7 Qxb7 22 h3 h6 23 Qh5 Bxg5! 24 fxg5 Nxg5 leaves White's kng fataly expose d. 19  h6 Exploitng the pn and effectively ndercuttng White's attackng ambitions. Now despite White's great efforts, he fals to reach h7 and the square remans tenantless. His attack bogs down, while Black's zoomng conterattack slalom's effortlessly to the fish ne.

20 Q5 Qxc3 21 Rd1 The rook must guard the first rak. 21 xa7?? Qe + 22 Kg2 Nd2+ mates. 21   ex 4 22 B2 22 Bxd4 Qxc2 23 gxh6 Nf6! overloads White's queen, who cnot smultaneously cover the d-rook nd the mate threat at g2. 22  Qc6 The queen's naggng wears on White's kng.

23 gx6

 ( ): Pressre monts on White's kng, who sits alone n his provisional rule, soon destned to lose all power. How did Botv force mate?

 The Red Sea parts. Double attack. The threats on g2 nd h3 are nanswerable. 23  Ng5! White can no longer conceal the ugly blemishes along the h-a diagonal. The kight offers a precious gft: his fe. Now White can do nothng more than watch hope dish and recede nto the horzon. 24 Rg3 Qh+ 25 Kf2 Ne4+ 0-1 Game 7 MTal-MBotvk World Championship (9th matchgame), Moscow 1960 Caro-Kann Defence

ne wonders just what heights Tal would have scaled f his nd could have been transplnted nto a healthy body, and f he had preferred soyilk tea lattes to vodka. Tal's rearkable chroicle on his 1960 match with Botv is probably the best match book ever written. As a kid I eagerly pored through Tal's nsights. My oly regret is I would be a stronger player to day f I had spent more time n my misspent youth studyng games ike this (rather than chewng on grass while gazng upon the Cnadian, ale-colored srise n a daydream each morng). Tal's book dffers from others. Instead of dong out reams of data, Tal offers the reader a gimpse nto his personal feelngs, thoughts, agitation, joys nd

disappontments of a direct participant n the combat."

1 e4 c6 The Caro-Kan is a sensible choice aganst a computer-ike calculator opponent. 2 4 5 3 Nc3 xe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 N1e2!?

: Why block n the f-bishop?

 The block is oly temporary snce White plans to move the kight to f4, from where: 1. It may pick off Black's ight-squared bishop. 2. It may assist n harassng the bishop frther with h2-h4-h5. 3. It eyes e6, a temptng target for a sacrfice. 6   Nf6 7 h4 The imediate 7 Nf4 is more common, as n the later game M.Tal-M.Botvk, SSR Spartakiad 1964, which contnued 7 ... e5 (probably better than 7 ... e6 8 h4 Bd6 9 h5) 8 Nxg6 hxg6 9 dxe5 Qa5+ 10 Bd2 Qxe5+ 11 Qe2 Qxe2+ 12 Bxe2, when White's bishop par ay be offset by Black's active pieces and open h fle. 7  h6 Hrryng to give his bishop some air.

8 Nf4 B7 9 Bc4 A sac on e6 beckons. Tal had the equal fortnemisfortne of beng tantaized with ideas most others regarded as preposterous. His healthy obsession with the e6-square is destned to be the central node to his subsequent misfortnes n this game. 9  e6 10 -!?

: Isn't this rather dngerous for White after he pushed forward his h-pawn?

 It is, but Tal smply looks to make trouble everywhere he can fnd it. White puts his own kng n possible futre risk n order to deploy a rook rapidly to e, where it adds a third agitator upon the e6square. 10 c3 may be a safer corse, but with the downside of reduced complications as wel. After 10 ... Nbd7 11 Qf3 Nd5 12 Ngh5 N7f6 13 Bd2 Nxf4 14 Bxf4 Nxh5 15 Qxh5 Bd6, Black achieved a re asonable position n .Korneev-A.iazantsev, Portuguese Team Championship 2007. 10  B6  Nxe6!?

Such a bold assertion also comes with great peril to the aggressor. This is vntage Tal, whose sacs often exude from the recipient, that credlous, chld-lke, it-can't-possibly-work/ feel to them, yet through his mysteriously awesome power of tactics nd deceit, they so often did work. In situations ike this, his complsion to sacrfice somethng  anythng!  was amost Pavlovian. Here he erupts n a typicaly grand gestre, displayng his felong nclnation to gamble, even when the sart money wold bet aganst his chances. A machne can't be designed to play lke this, snce a non-sentient chess program is ncapable of Tal's implsive spontaneity. (Well, okay, not this time. His move was home prep!) Perhaps a computer's fatal flaw is its nablity to dream.

: Holy mother of Alekhne!?

 Technicaly, that is not a question. But I share yor sentments. Clearly, we are dealng with a man who fails to hold material thngs dear . Sond sacs were never Tal's strong sit! If he sac' ed, it was often fishy, yet they worked so terrfyngly often aganst his befuddled opponents. But not this tme. Although the sac was pre-game prep, to the Tal camp's disay Botvk ade his next few moves very qickly, cofrmng that he had foreseen it and worked it out the defensive formla at home! Such was the level of Botvk's legendary preparation, that he was ready and on the alert for moves which had never been tried befo re! Tal writes: Botv usually won by gettng his opponent n a vice-lke grip without givng hm any respite./ So Tal's successfl strategy throughout the match was to disrupt the natral strategic flow with arrng anarchy. He contnues: r (Tal's analytical camp) iscalcation was that this time we had somewhat nderestiated the phenomenal analytical powers of Botvk, even assng that this sacrfice wold be nexpected (which it wasn't!)./ 11  e6 12 Bxe6 Qc7

: What compensation does White get for the piece?

 Suddely, Black's kng gives rise to feelngs of consternation upon viewng the approachng hostles. White's comng attack, for now, is an amorphous creature, devoid of shape, whose presence we merely sense, yet canot distngish from its envroment. The lowbrow attack, upon closer nspection, turns out to be closer to a iddlebrow one  not so easy to refute. Tal hmsef responds: 1. His bishop on e6 prevents Black from castlng on either side. 2. The open e-fle lkewise cofrms the fact that Black wil scarcely be able to castle at all n this game  he wil not have the tme and therefore it might take several moves to get his rooks nto the game. 3. If the white-squared bishops are exchanged, then the white kight goes to f5, from where it wil be able to put dangerous pressure on Black. If Black prefers to elmnate the kight, givng up his black squared bishop for it, he wil have catastrophic weakness on the dark squares and White's queen's bishop wll take up a very dangerous position on f4." Tal gives the ne 12 ... Bxg3 13 fxg3 Bg8 14 Qe!! Qe7 15 Bc8! with the ntention to completely plnder the queenside ." 13 Re1 Nb7! Darng White to lanch a discovery. 14 Bg8+ Swappng off Black's best defender so as to weaken f5. 14  Kf8! The old kng capers about with the agility of a much yonger an. Tal laments that Botvk n this game, tme and tme agan nearthed the best moves, even nder terrible duress, claing that The position of the kng on d8 would have been much worse."

15 Bx7 Rx7 16 Nf5

 (  ): Tal's camp exaned this position and ncorrectly evaluated it as rosy, snce White's kight is there to stay on f5 and the h7-rook is badly out of position." How did Botv solve both these problems with just one little move?

 16  g6!! Hey, Tal said White's kight is there to stay!" This startng idea had apparently been overlooked by camp Tal. Botv nearths a briliant defensive plan which dampens and constricts the flow of White's attackitiative.

: Why did Botvk hand over his h-pawn, and with check to boot?

 By retrng the pawn, Black solves two ajor headaches and displays the apparent, imperanent, evaporative natre of White's itiative: 1. He elmnates that hateful kight from f5. 2. He mmediately activates the sleepng h7-rook. Very soon, the attack which once was, quickly fades and passes fr om memory  and Botv's kng, by what appears to be some mystical nadvertence, passes through his o rdeal nscathed. 17 Bxh6+ Kg8 18 Nx6 Qx6 Tal ponts out the deviish trap 18 ... xh6 19 Re6 xh4 20 Qd3?? (20 g3! is correct, when White's super active pieces compensate for his aterial deficit) 20 ... Nf8! 21 xf6 Qh7!.

19 Bg5 Re7 White's attack grows cold n a condition of dismal grander and is no more. Just look at the dfference: Black's pieces emerge, whle White's mope about n desultory fashion. The extra kight is superior to White's three pawns. 20 Q3 Kg7 21 Qg3?

·



A felong physical aberration or defority always seems normal to the one gazng n a irror. The queen attempts to dance away from an earlier proise to deliver checkmate, but I suppose n an econoic depression there are few buyers and a glut of people desperate to sel. Tal, n a hrry to sell, nexpicably endeavors to go on strike aganst logic, without an iota of basis behnd the move. The ruly queen flagrantly disobeys societal laws, violatng the prnciple: Avoid trades when behind in material and structure. Tal short-crcuits and ilogicaly offers to swap queens (which Black wants) and damages his structre n dong so (which Black wnts !) . The move strikes one as decidedly out of sync with the position's requirements.

: How can a player as formidable as Tal make so lame a decision?

 Tal saw ghosts nd malicious sprits hoverng arond his own kng, and felt he had to take queens off the board, ike it or not. He cofessed: nquestionably the weakest move of the atch. I frakly thought it was bad, but somehow al the other contnuations were worse." World Championship matches have their own rules, and the mnd of geius is n fathomable envroment of wonderful secrets and strange motivations. The nconceivable stran, for so long a time period, wears down a player's nervous system. Do you remember Bobby's Fischer's nexplicable 29 ... Bxh2??, sef-trappng his bishop n his frst atch game aganst Spassky? Tal ponts out the plan which puts up the most resistance: 21 f4 Rae 22 Re!, but even here my spider senses tel me Black can ravel and consolidate. All the same, this is the one lucid hope to rehablitate White's flaggng tiative. 21  Rxe1+ 22 Rxe1 Qxg3 Thaks!

23 fxg3 White's kngside pawn ajority has been fatally comproised. 23   Rf8! Ruthless accracy. Botvk blocks White's kng access to the centre, snce ... Ne4+ would be fatal.

24 c4 Ng4 25 d5 Passive defence fals; e.g. 2 Re4 Ndf6 26 f4 Re. 25   cx5 26 cx5 N  27 6 R

·



The overextended d-pawn, a soitary anoaly among Black's mltitude, soon falls. White's pawn has rn agrond and Black ntends to systematically destroy it."

: If this is the case, then why did Tal push it to a vnerable sqare.

 Tal probably decided to jettison the pawn n the hope of distractng Black's rook, so as to activate his own kng.

28 Ri Rd7 29 R7 Tal noted to his dismay that the combnation 29 Bf4 Nd5 ! 30 Rc7?? Nxc7 31 dxc7 Rd1 mate ends on a disal note for White. 29  Kf7 Botv's kng dutfly rns off to fetch the d6-pawn with Golden Retrieverish zeal.

30 Bxf6 Tal jmps on the chance to free his kng, but oly at the cost of another swap. 30  Nxf6 31 K Ke6 32 Rx7! White's sim hope of a draw is to remove r ooks from the board. 32   Kx7 33 Kf3 Kx6 No desired object is too expensive for a thief, who views it all as free. 3 Kf4 Ke6 35 g4 N5+ 36 Ke4 N+ 37 Kf4 N5+ In orde r to gan advantage over Botvk's felong, hated enemy  his clock.

38 Ke4 Nb4 The imediate 38 ... g5! is very strong too.

39 a3 Tal felt his move was a blnder:  ... the pawns on a3 and b2 are wonderfl targets for Black's kight." 39 a4! wold signficntly ncrease Black's dfficty n realzng his advantage." But I th everythng loses at this pont.

: What wold Black's wng technique be after 39 a4 then?

 39 . . . Nc6 40 h5 g5 41 h6 Kf6 42 Kd5 Nb4+ 43 Kd6 Nd3 44 b3 a 45 Kc7 Nc5 46 K6 Nxb3 and f 47 Kxb7 Nc5+ removes White's last chance. 39  Nc6 40 h5 g5! Pnciple The piece up side should avoid pan exchanges.

41 6 A artyr tends to roanticize his ow n sfferng. The fnal hope  n ord er to deflect Black's kng nd allow White's entry to the queenside. But as it trns out, the lavish expenditre n van. 41  Kf6 42 K5 Kg6 43 Ke6 43 Kd6 Na5 44 Kc7 b5 45 Kb8 Nc4 46 Kxa7 Nxb2 47 Ka6 Nc4 is crtans for White. 43  Na5 4 a4 Nb3 Tal complans: This wold have been of some use to White f he had correctly played it on the 39th move!" But I don't thk it wold have mattered. 45 K6 a5 46 K5 K xh6 47 Kc4 Nc1 48 Kb5 N3 49 b3 Nc1 50 Kxa5 Nxb3+ 51 Kb4 r 51 Kb6 Kg6 nd captre of b7 is agan met by ... Nc5+. 51   Nc1 52 Kc3 Kg6 53 Kc2 Ne2 54 K3 Nc1+ 55 Kc2 Ne2 56 K3 Nf4+ 57 Kc4 Kf6 58 g3 Ne2 0-1





After 59 Kc5 Nxg3 60 K6 (White's kng, footsore and weary from his ailess meanderngs, now lays down to rest  he can oly sigh and muse upon those halcyon pre-sac days when material was even and his kngdom at peace and content) 60 ... Ne4! 61 a (once agan choppng on b7 is met by that anoyng kight check on c) 61 . . . Nd2 62 Kxb7 Nb3 63 a6 Nc5+ 64 K6 Nxa6 65 Kxa6 Ke5, Black wns easily. Game 8 JLttlewoo-MBotvk Hastngs 196162 icilian Dence 1 e4 c5 2 Nf3 6 3 4 cx4 4 Nx4 Nf6 5 Nc3 g6 Botv was an early practitioner of the Dragon, datng back to the 1930s.

6 Be3 Bg7 7 £3 a6





The Dragodorf, a hybrid ne which combnes the mean-spiritedness of the Dragon with the deceit of the Najdorf.

8 Bc4 Littlewoo d procee ds n ugoslav formation.

: Can White hold back the bishop's development to avoid the loss of tempo?

 es, White can refran from Bc4 as wel, although his control over d reduces. A.Shrov S.K.Wiliams, British League 2004, saw 8 Qd2 Nbd7 9 0-0-0 b 10 g4 Bb7 11 g Nh 12 Nce2 Nb6 13 Ng3 Nxg3 14 hxg3 d!, when Black took advantage of White's neglect of d and eforced his thematic break. Principle Counter in the centre when attacked on the wing.    b5 9 Bb3 Bb7 10 Q2 Nb7 11 0-0-0 Nc5 Botv sets his sights upon the powerful ight-squared bishop.

12 Kb1 After 12 Bh6 Bxh6! 13 Qxh6 e! (Black switches to full Najdorf mode) 14 Nde2 Nxb3+ 15 axb3 Qe7 16 Rd2 0-0-0 Black equaized n S.Feller-T.Gelashvii, anty-Mansiysk lympiad 2010. 12   N xb3 13 cxb3 : White's last move looks pretty strange to me. Why would he capture away from the centre?

 Not so strange, snce every game n my database has White akng the same capture n this position! There is a defensive prnciple: Capture a om the centre  you sense that your king is in danger.

: In this case, how does that help White defend his kng?

 Now f need be, White has the option of chalengng Black on the c-fle, which leads to swaps. 13   0-0!? Botv was never afraid of an opposite wng attack n his lfe. The alternative is to stall with 13 ... h!? as n L.Cernousek-K.Shanava, lomouc 2006, leavng Black's kng n the centre as long as possible.

14 Bh6 Bxh6! Deflectng White's queen from the centre and correctly judgng that she wll not fict daage on h6.

15 Qx6 The queen transmits secret orders to the front, but what she doesn't realze is that her phone l is not secured.

15   b4! 16 e5?! ften, fncy doesn't equate with best. 16 Nd5 is stronger and more natral; eg 16 ... Bxd5 17 exd5 Qd7 18 h4! Nxd5 19 h5 Nf6 20 g4 Rac8 21 h2 Rc5.

: Isn't Black about to get mated?

 White's attack looks scary but f you play arond with the computer n this position you see that Black defends. Let's go deeper nto the analysis: 22 g5 Nxh5 23 xh5 gxh5 24 h e5 25 Rxh5 f6 26 Ne6! f7 27 Nxc5 dxc5 28 gxf6 Qd + 29 Qc1 Qxc1+ 30 Kxc1 Rxf6 31 xe5 xf3 32 xc5 f6 with a sight edge to Black, who can stll dream about big thngs to come with his h-pawn. 16  N7! Not 16 ... bxc3? 17 exf6 exf6 18 bxc3 when White's kight donates its bishop conterpart on b7, who may be labeled a sub-species, a cheap facsmle, when compared to his more evolved brother on d4.

17 4 O 17 e6, Black defends with 17 . . . bxc3 18 exd7 Qxd7 19 h4 e5! .





The dfferences between the parties reaches irreconclable levels and war is inent_ Littlewood goes for it with a dangerous piece sac which would probably have worked aganst an opponent who didn't calculate ike Hdini_

 ( ): Can we take the c3-kight and live, or is there a better move?

 Black gets away with the piece grab_ White believes it is Christas Day_ He is wrong  it is udgment day_ 17   bxc3! Thnks, you shouldn't have! Botvk worked out every potentiality and came to the correct conclusion that they all fel short for his opponent_ Even though White's sac fails to hold up nder analytical scrutny, this doesn't mean it isn't dangerous to a human laborng aganst the clock, and also perhaps paranoia_

18 5

: Isn't Black just gettng ated! ? How does he defend h7?

 I'm afraid yor authorzed outbrst cals for a defensive exercise! Attackers approach Black's kng with open ntent. It does ndeed look as though the defence reaches satration levels, nable to take the pondng a nute longer as rival factions struggle for hegemony over h7. However, hidden n the secret weaves of the position hides the correct defensive plan, scribed n the crsive of a dead language. Cn you extricate it and discover the position's essential core?

 (  ): Black to play and not get mated!

 Clear f6 for his knight. 18  xe5! 19 hxg6 N ut of nowhere, Jack ateriaizes from his box.

20 bxc3 If a child gets nabbed red-handed while perpetratng ischief, her implse is to try and ie her way out, no atter how obviously glty.

: Isn't White's last move somewhat accomodatng?

 I agree that his overreaction, which is disproportionate to the offendng provocation, converts that which was once a drama nto a hastly rewritten farce. But it's hard for me to give White a question mark for a move n a position where every other attempt fails as wel.

: What? It looks ike White just went crazy n a position where he must have three or for ways to force ate.

 At first glance, it does ndeed look lke there shld be a ate for White, but there just isn't. It's as i the secret formla, written on a piece of scrap paper, gets nadvertently washed n the pocket of a pair of jeans, and is now hopelessly bleached nd readable, the discovery lost forever. Littlewood anaged

to achieve his fantasia of attackng possiblities, yet not one of them did hm a bit of good. The humn bran is capable of absorbng oly so any fragments of data before it overloads nd short circuits  but not Botv's bran apparently! Behold, Botv's co mp-lke calculation powers. Every lne I tried aganst the computers failed miserably for White:

a) 20 gxh7+ 8 21 Nc 2 Nd5 22 Ne3 Qd7 23 g4 Qe6 and Black covers al hostle ntent. b) 20 g7 Re8 21 Nf5 is met by the cng shot 21 . . . c2+! 22 Kxc2 Qc8+ !, which picks off the would-be hero on f5. c) 20 gxf7+ Kxf7! covers the fork threat. And remember, any Nf5 is met by the crushng . . . c3-c2+! trick. d) 20 Nf5 c2+! (is this theme begng to have a famliar rng to it?) 21 Kxc2 Qc8+! etc. 20  ex4 Wealth concentrates n the greedy, outstretched hands of a sngle ndividual (Botvk), whle the rest are left to starve. 21 gxh7+ Now White's h7-pawn chokes off his own attack and the exhausted yet triumphant black kng basks n glowng languor on h8, safe as can be. 21  Kh8 22 Rx4 Qa5 23 Qe3? White's attack proved to be merely a temporary leaven, which deflates to its new and fortnate state. The two sides are now governed by opposite motivations: Botv's to prsue and Littlewood's to srvive. The queen sighs and retreats, the way a teenage girl rolls her eyes at her mother when ordered to perform a distasteful chore arond the house. Retreatng the queen to d2 was better, but maybe resignng was White's best. Not such an encoragng sign. White is down two pieces and n the middle of his attack he must take time out for a defensive move! 23  N5 Forkng e3 and c3. 24 Q2 Nxc3+ 25 Ka1 Ra8 26 Rc1

 ( ): Calculate 26 . . . Qxa2+. Is it playable?

 It certay is. Black simplfies down to a trivially won endgame. 26  Qxa2+! The controlng shareholder gets the fnal say.

27 Qxa2 Nxa2 The white kng coments: While I'm not physicaly n pan, the loss of digity stil realy hrts!/ 28 xd8 xd8 0-1 Game 9 MBok-RJFscher Varna lympiad 1962 Grnfeld Defence

The moment had arrived. The World Champion Mikhal Botv faced the futre champion n what became a titnic struggle. In a strange way the game proved nothng, snce Fischer was oly 19 years old and not yet ischer, while Botv was a tad past his prme (nd yet stil world champion!) nd no longer otvinnik 1 c4 g6 2 4 N 3 Nc3 5!? Fischer said he ntuitively switched from his noral Kng's Indian to the Grfeld because of the gnt n his [Botvik's] eye./ He just had a fny feelng Botv would be ultra-prepared for hm n the KID. But as it trns out, Botv had come armed for the Grfeld as well.

4 Nf3 Bg7 5 Qb3 Smyslov's specialty. 5  xc4 6 Qxc4

: What is the benefit? Isn't White's queen vunerable to tempo loss?

 It is, but White gets somethng n retrn: Black is deied his ... Nxd5 nd ... Nxc3 freeng exchange, which is impor tant n the Grfeld, snce Black remans more cramped than normal. 6  0-0 7 e4 Bg4 8 Be3 : White isn't concerned about . . . Bxf3 - ?

 Not really  he bags the bishop pair and, fter g2xf3, strengthens e4 as wel. The dowside, of corse, is that White's shelter on the kngside grows somewhat compromised. 8  Nf7 : Why move the already developed kight?

 Black hopes to engneer ... c7-c5 or ... e7-e5. By movng his f6-kight to d7, he lays the grondwork for either pawn break.

9 Be2 9 Qb3, 9 Rd and even 9 0-0-0 are played here as wel. 9  Nc6 Fischer wisely veers from 9 ... Nb6 10 Qc5 c6 11 Rd N8d7 12 Qa5 e5! which alowed Black dynamic equaity n M.Botv-V.Smyslov, World Championship (4th atchgame) Moscow 1958. However, the odds were close to 100% that Botv was ready with an mprovement n this variation.

10 Rdl Nb6 11 Qc The queen must contnue to cradle her tender d4-pawn. 11  Q6!

12 3 12 Qxd6 cxd6 allows Black a very decent Prc-ike position, where White's extra space ceases to be such a great brden on Black with queens off the board. 12  Bxf3 13 gxf3 : Why not recaptre with the bishop?

 Dong so weakens the c4-square. Black looks good fter 13 Bxf3 Qxc5 14 dxc5 Nc4 15 B fd8, when he leads n development. 13  Rf8 Botv would welcome 13 ... Qxc5 14 dxc5 Bxc3+ 15 bxc3 Na4 16 Kd2!, when his bishops ight later have a say.

14 d Ne 1 Nb Qf6

Fischer gave himsef an exclam for this move and called 15 ... Qxc weak". Kasparov disagreed, offerng the lne 16 Bxc c6 17 Nc7 (17 Nxa7? is met by 17 . . . Na4! ) 17 . . . Rab8 18 Bxe7 Rd7 19 d6 Nec4 and Black should be okay. 16 f4 Ne7 17 e5? White's e-pawn gives offence to Black's queen and soon wishes he could nsay what he just said. Pieces and pawns entwne and grapple n desperation, neither side wilng to cede even an nch of hard-won territory. Botvk writes: When I was preparng to meet Smyslov, I, of corse, ade a thorough analysis of the Smyslov System n general and of the position n the diagram n particular!" Botvk had oly nalysed ... Qh4 nd ... Qf, concludng his analysis with a White advantage n both nes. But then Botvk adds: Alas, my opponent fond a third contnuation!"

 ( ): What did Fischer fnd n the position which took over the itiative, eventually won a pawn, nd threw Botvk badly off his prep?

 pon nventory, White notices an item issng from the shelves. verload. White's bishop can't simultaneously captre Black's queen and protect his own. 17  Qxf4! Fischer comes up with a strong theoretical novelty. The black queen offers her services with nsncere solicitude, as she arrives on f4 with good tidngs of cofort and joy to all. Botv laments: I had missed what Fischer had fond with the greatest of ease at the board. The reader can guess that my equaiity was wrecked." All the same, the by-blow ficts damage, yet fals to kil.

18 Bxf4 Black's pont is that 18 Qxb6?? fals to co llect war dividends because of 18 . . . Qe4! and White loses material, snce 19 f3 is completely nplayable due to 19 ... Qh4+ 20 Bf2 Qb4+, pickng off the queen. 18   Nxc5 19 Nxc7 Rac8 20 6 Botv's game borders on overextension, but he has no choice snce 20 Nb? xd is terrible for White. 20  ex6 21 ex6 The d6-pawn ay have falen to his knees, yet reans a dangerous foe, a wonded gladiator, brought low but stil managng to clutch sword and shield n each hand.

21  Bxb2 22 0-0 Nb7 Botv preferred 22 ... Ncd7.

: It looks to me lke Black is all tied up defendng aganst the crampng effects of the d-pawn. White owns the bishop par n an open position as well. Don't these two factors constitute full compensation for a pawn here?

 Perhaps some compensation but not ful. Black is a healthy pawn up with very active pieces to boot. White's d-pawn remans frmly blockaded and his compensatng activity represents a mere abstraction snce there doesn't exist an accompanyng plan to ncrease it. 23 R5 b6 24 Bf3?! Botv's forces creep forward with enticng lassitude. There is nothng for the bishop to do here," writes Botvk, nstead suggestng 24 Bc4!, which keeps an eye on e6 nd f7. White may then extract full compensation for his missng pawn. 24  Ne6! Fischer seizes upon the naccuracy. 25 Nxe6? Botv, stil shaken by Fischer's superior analysis n the openng stage, begns to fal apart. 25 Bg3 was correct. 25   e6 26 R3 N c5 27 Re3 Probably played with deep regret. White must hand over his pride and hope on d6 snce the exigencies of his very survival deand it. 27  e5! Dual purpose: Elmnatng White's bishop pair and the deeply entrenched d6-pawn n one swoop.

28 Bxe5 Bxe5 29 Rxe5 Rxd6 The orphaned d6-isolai gets tossed o ut nto the cold.

30 Re7 Rd7 31 Rxd7 Nxd7





A disal pictre. White deals with a distressng aalgam of woes: 1. Botv fnds hmsef a pawn down. 2. The landscape of White's structre  he nrses three isolated pawns  is the residue of a ightare upon awakeng. 3. Makng matters worse, Botvk must defend this awfl endng aganst a player who, n the writer's opnion, was the second strongest endgame player of al tme. Fortnately for Botv, Fischer was oly 19 years old at the tme, and hadn't yet reached the peak of his endgame sklls.

32 Bg4 Rc7 33 Re1 White ight consider 33 Bxd7 xd7, trnsferrng to a rook and pawn endng. But even here he nrses mltiple, chroic, structral aiments. 33  Kf7 34 Kg2 Nc5 35 Re3 Re7 36 Rf3+ Kg7 37 Rc3 Re4 38 B1 R4 Possibly naccrate, allowng White's bishop to its best post on c2.

: How can Black prevent this?

 By playng 38 . . . Re!, when 39 Bc2?! waks nto a pn after 39 ... Rd. 39 Bc2! Kf6 40 Kf3

 (  ): ek out a step-by-step pln for Black to convert to a wn.

40  Kg5?! Although White's prognosis looks ismal, there now appears a m hope. Fischer goes off on the wrong track. Botvk answers the exercise, with an ae nti-Fischer barb as a bonus.  /IAn engame specialist of the class of Capablanca or Smyslov woul have mmeiately transferre his kng to 6, efenng his kight, after which the avance of the queensie pawns woul have ecie the outcome. 41 Kg3 N e4+?! A violation of the prnciple: Rook and pan endings are notoriously hard to win a pawn up. Another poor ecision n another anti-Fischer statement to come: /The efects n the character of my opponent begn to tel. Reckong that the position was easly won, he was ngry with me for playng on, an n his fervour, alreay after the time control he makes a rash ecision. One sometmes woners just how much tme Botvik nveste n ponerng some of the alice festerng withn his own psyche.

42 Bxe4 Now White gets a better version of a rook n pawn enng than the one mentione n the note to move 33 above. 42   Rxe4 43 Ra3 Re7 44 Rf3

 (  ): Fischer's kng is boxe n. Come up with a concrete plan to free him.

4  R7?!  The trouble with Shangri-La: i you leave, you age rapily an ie. n just the same way, Black's kng fns hmsef trappe withn his paraise. Here Fischer, nable to fn the solution, shoul strive to activate his kng, which can be accompishe by 44 . . . Kh6!, ntenng . . . Kg7 n . . . f7. 45 a4!

: hat is the pont of this move, which just looks lke it weakens?

 Botvk hopes, with a4-a5 at some stage, to swap one set of pawns on the queensie, which woul ease his efensive task. Someone aske Botvik at breakfast what he thought his chances were. He sai without lookng up, "Nichia  raw! Apparently Fischer felt the position was a straightforwar wn for Black an han't worke too har on the ajorment. Botv, on the other han, refuse to give up, an labore on the position ntl 5:30 a.m. the next morng. Apparently it was time well spent, snce his ream team  Boleslavsky, Spassky, Geler, Keres n Fran  toing together with their leaer, nearthe an ncreibly hien rawng iea. 45  R5! The seale move an, as it trns out, the strongest n the position. Fischer, r on extation, can just taste it now. Botvk looks buste an Fischer's victory seems to be n plan sight. hat Fischer fale to factor n was Botvik's vast atabase of engame nerstanng, which exceee even that of Fischer's at the tme. 46 Rf7 Ra5 47 Rxh7!!

·



A stg iea, iscovere by Geller. Magic oozes from the wzar's ara, so that canle flames bow n respect when he enters the room. Corresponngly, Black's ark powers noticeably begn to abate. Apparently, Fischer ha also analyse the move but misassesse it. White neatly fns a metho of fyng aggregate strategic troubles an eventually issolvng them. Soon, Black's once-proisng futre of a full pont is quashe on appeal.

:  on't get it. This looks resignable for Botv's sie now. Why on earth woul White allow Black two conecte passe pawns on the queensie?

 White plays on the prnciple: Ke your rook active at all costs in rook and pawn endings. White correctly assesse that passive efence woul lose. Play on to iscover the rawng iea. nstea, 47 f4? f5! 48 R4 f7! alows Black's kng out of his box an nto the fight on the queensie. White can't save the game here. 47   Rxa4 48 h4+! More accrate than 48 f4+, the move Fischer ha concentrate on n his analysis. 48  Kf5 48 . . . Kf6 49 Rb7! (the key move, which raaticaly slows own Black's pawns) 49 . . . Ra5 50 Kg4 b5 51 f4 a6 52 Rb6+ Kf7 53 Rb7+ also leas to a raw, espite Black's prze pair of passers on the queensie. 49 Rf7+ Ke5 50 Rg7 Ra1 50 ... Kf6 51 Rb7! trnsposes to the previous note.

51 Kf3 Botv, even with virtually zero sleep, isn't about to fal for 51 xg6?? Rg +, wng the rook. 51  b5?! Botv offers lengthy analysis, clamng that 51 . . . K4! is more challengng to White, but stll mantans that he raws, whereas Fischer vigorously isputes the assessment an claims a wn for Black! Kasparov backs Botvk's rawng clam, so take yor pick n beieve who you lke!

52 h5!! A sweet fnal bar n a rousng symphony. The once nert figre from a two-imensional pantng, through strange magic, takes birth an steps out nto or three-imensional worl. Botv wrote, with obvious aicious glee: /Here my opponent trne pale . . .  The profon pont of White' s 47th move, which forces a pair of rook pawns on the a-n h-fles, is apparently a rawn position! Supposely, after this move the titans create more rama. Botv wake up to the Russian captan Abramov an repeate /Nichia! most certanly ntene to rattle Fischer. The always parnoi Bobby overhear, went balistic, an accuse the Russians of collusion an cheatng to the arbiter. Black is faster n the race after the ro te 5 2 xg6? b4. 52  Ra3+ 53 Kg2 gxh5 54 Rg5+ K6 55 Rxb5 h4 56 f4 Kc6 57 Rb8 h3+ 58 Kh2 a5 59  Kc7 Black fns that a lone pawn, the paltry sm of his wealth, is nsfficient for victory. 60 Rb5 K6 61 f6 Ke6 62 Rb6+ Kf7 Black's kng contnues to nrse grievances aganst his tormentor but is powerless to o more than merely voice a complant. 63 Ra6 Kg6 64 Rc6 a4 65 Ra6 K 66 Rc6 R3 67 Ra6 a3 Black attempts to her cats, prong the rly h-an a-pawns forwar to nowhere n particar. 68 Kg1 /-2

Black has no way to ake progress. The fny thng is that the computers get foole here. Hdini misevaluates it at  2.01, wng for Black, when the reaity is a ea raw.  playe it out aganst Hdini, an it coln't o a thng an allowe an easy raw. t was reporte that Fischer left the tornament hal n tears. Botvk salte the won frther when he wrote of Fischer, who he clearly espise: /Success n chess is ecie not oly by talent, but also by other quaities, nclung the character of a player.  He contnue later with /I n Fischer' s character was always clearly naequate, as the reaer wil probably agree, after playng through or game. Ouch! Kasparov, not a Fischer fan either, ple on, writng about the misse wn, /I • • • Bobby lost part of his halo • • •  As revenge, Fischer, n an article two years later, offere his list of the top ten greatest players of al

tme, with Botv noticeably absent from the list! Ouch agan! Game 0 P Beko-M Botvk Monte Carlo 1968 English Opening 1 c4 g6 2 g3 Bg7 3 Bg2 e5 4 Nc3 Ne7 Toay, theory frowns upon e7 as a sub-par post for the kight, although  stil see recent granmaster games with it n the atabase. Better to toss n . . . f7-f5 earier an then follow with . . . Nf6; e.g. 4 ... Nc6 5 3 f5 6 e4 Nf6 7 Nge2 0-0 8 0-0 6, Y.Balashov-S.Dvorys, Eropen Cup, Cheliabnsk 1991.

5 e4 Beko, perhaps n an attempt to mess with his opponent's hea, playe Botv's own variation of the English Openg aganst its foner. Botvk ntrouce the then startlng iea with reverse colors n his 1954 atch with Smyslov (which we exane n Chapter Five) an it so utterly iscofite his opponent that it nuce Smyslov to abanon the Close Siciian altogether as White.

: Doe sn't it create a gapng hole on 4?

 t oes nee, but Botv reaize that the hole's ngers are more appearance than reaity. White ay later simply captre on 4 an plug the hole with a black pawn. Meanwhile, White receives far greater central control than n other variations of the English versus ng's nian. 5  6 6 Nge2 Nbc6 7 3 f5 The imeiate 7 ... 0-0 is more common but it probably oesn't ake such a great fference here.

8 Nd5 M.Botv-T.V.Petrosian, USSR Team Championship 1966, saw 8 0-0 0-0 9 N5 8!? 10 Be3 Be6 11 Q2 Q7 12 Rae Rae8 13 f4 exf4 14 Nexf4 Bg8 15 Nxe7 Nxe7 16 Bh3! b5 17 b3 c 18 4!, when Botv hel a slight tiative n a very complex position.   0-0 f 8 .. . Nx5 9 cx5 Ne7 10 h4!? c6 (10 . . . h6 is a try, ntenng to bypass any h4-h5 with .. . g6-g5) 11 xc6 bxc6, N.Miezis-F.Velkhai, Geneva 1999, then  prefer White's position after 12 h5. 9 Be3 Be6 10 Q2 Q7 11 0-0 R! The f-fle is the optmal placement for Black's rooks.

12 Rae1 Raf8 13 f4

Beieve it or not, this move may actually be sightly naccrate.

: How can such a natral move possibly be naccrate?

 Kasparov himsef questions this move.  agree that it certay loo thematic n the position, but through one of Caissa's malicious vagaries it just isn't. As you well know, the chess goess loves to play practical jokes on her worshippers.  have fon mysef on the wrong en of her jokes for qite some time now.

: What else is there for White?

 He can temporize with a qieter pln. Kaspar ov suggests 13 f3! . White's st ats are much higher with this move; eg 13 ... 8 14 b3 Ng8 15 exf5! Bxf5 16 4! with an ege for White, J.Smejkal-A.Yusupov, German League 1992. 13  e4! 14 xe4 N!

: What is the iea behn this convolute-lookng move?

 This is actually Step 2 n an ncreibly eep strategic plan. Botvk explane his construct n the folowng stages: 1. Exchange on e4, nucng White to recaptre with a pawn, snce a bishop recaptre allows an eventual ... Nf5. 2. Exchange on f4, forcng White to once agan recaptre with a pawn, snce a piece recaptre hans over the e5-square to Black. 3. Swap off ight-square bishops via h3, weakeng White's e-pawn. 4. Shft rook (or rooks) to the e-file. 5. This n trn nuces Ng3, to cover the weakene e4-pawn. 6. Follow with ... h7-h5! an ... h6-h4, nermng the efener of e4, after which White ay have great fficlties efenng the pawn.

: This plan sons absolutely logical an strong but how i he work all of this out over the boar!?  wol never be able to o this kn of plang.

 Because he was Botvk! His games, especially ones nvolvng strategic plang, are rearkable for their iron clarity, as well as ther ncreible epth. As  mentione n the ntrouction to the book, Botvik, Morphy, Capablanca, Fischer an Karpov ( have a feeng Carlsen may soon be ae to this list) were the omnant strategists of their generation (or any generation for that matter!) an were, n my opion, the greatest positioal players of al time.  remember gong over this game when  was aron 10 years ol. Botv's explanation of his n-epth plan blew my ki n! t was actualy a revelation to me at the time.  realze that epth attere, an to mprove my game  ha to get past vlgar, two-move cheapos (my ealy arsenal back then!) n actualy work out plans schematically, not ust athematically.

15 c5 Beko softens up Black's centre, meetng a wng attack with a central conter. Kasparov also mentions 15 b4. 15  Bh3! Step 3: Swap off ight-sqare bishops. 16  Bxg2 17 Kxg2 exf4! Step 4: Weaken e4. 18 gxf4 Re8! Step 5: Pressre e4, to nuce Ng3.

19 N3

19  h5?! Step 6: Unermne the efener of e4.

: What? Why i you criticze Botv's logical move, which was al part of his plan?

 Kasparov gave Botv's move an exclam an Botvk gave it the seal of approval too, but it is istime. 19 ... N6e7 is better here.

: You, a pny ittle M, are to chalenge the assessment of two Worl Champions?

 Wel, it oes appear a bit presmptuous on yor writer's part, but n my efence, there are mitigatng factors: Botvk in't have access to a computer. Kasparov i but he wrote his notations aron 200203, usng computers much weaker than toay's moels. The Houdini an ritz programs of 2013 are several levels stronger n apparently fon a hole n Bot vk's move orer.

Beieve me, ten years from now i you put this analysis n the top program of 2023, you wil nearth a mlion mprovements as well. This is why ( believe, at least) actual analysis n a chess book isn't al that importnt (!), snce the "best move or lne is n constant flx, ue to the ncreasng strength of chess programs. y the prose, assessments, opons, n verbal explaations of the thought process wil be of real use n a few years to come. 20 b5?! Hdini's impossible-to-fn refutation of this classic game rns: 20 f5! h4 21 fxg6! ! (the grng, re boie mp contnues to poke an pro  with his trient from g6) 21 . . . xf1 22 xf1! hxg3 23 Q1 ! (threateng Qh5, which n trn forces Black to retrn his har-earne gans) 23 ... Ne7 (23 ... f?? 24 Qb3! !  25 xf+ B xf 26 Q1 ! an it's game over !) 24 Qh5 Nxg6 25 Qxg6 f 26 Rxf+ Kxf 27 cx6 cx6 2 hxg3 with a wng position for White. Dang. 1 smultaneously loathelove these hateful yet allrng chess achnes. They have the awful power to embolen the weak an cofuse (saly, yor writer must be nclue n this fortnate group) an elue us nto beievng we play better than Capa, Botvk or Fischer, when n reaity the computer performs al the work while we are just along for the rie! 20  N6e7 21 !





Beko, sensng strategic vunerablity from his sie, goes for it. 21  h4 22 g6 White's hope is that the g6-tumor grows so eeply an ntricately embee withn the vital orgns of Black's kng position, that it may be noperable. Hdini once agan nearths another mpossible yet playable ne n 22 Bh6! ! (the orer to execute Black's kng is eivere by papal seal on h6  fter this move, White's nenng attack is a magic treasre chest which, when emptie, iraculously repleishes itsef with gol an jewels) 22 . . . Nx5 (not 22 . . . hxg3? 23 Bxg7 Nx5 24 c6! !  power  oesn't always equate to a numerical avantage i one sie has access to a secret weapon to offset the superior numbers  24 . . . bxc6 25 bxc6 Qxc6 26 ex5 Qb5 27 f6 with a ecisive attack) 23 Bxg7 Kxg7 24 fxg6! xf1 25 xf1 Qg4 (25 ... Nf6?? loses on the spot to 26 Qg5!) 26 Qx5 Qxg6 27 f3! with ragng complications an even a tny ege for White, accorng to Hdini. 22   Rx 23 Rx! t is too late to back own now with 23 Nxf1?? Qg4+. 23   hx g3 24 R!





Very few players woul srvive Botvk's sie aganst a top-level GM. Black's efensive hrles are: 1. White's pieces swarm over the kngsie. t feels to Black's kng ike his guar callously abanone him n his tme of nee. 2. The ark squares are near collapse aron the black kng. 3. White's rook nfltrate f7 an pns the e7-kight, whle his lazy, feckless brother contnues to longe on c. With this kn of raw ata assaultng Black, it is har not to istl it nto paic. Rearkably, Botvk manages to stay calm n form efensive coherence from cofusion. 24  Be5! Chastity, poverty, humble submission, are the holy vows the bishop once took n now eeply regrets. The ark squares must be chalenge at al costs. 24 ... Qe6?? 25 xg7+! Kxg7 26 B4+ wns. 25 B4 Qg4!





Kasparov writes: "t's fficult to reconcile Black's contraictory, hostile attackng gestre from his overal passivity. Yet Botvk oes just that. Black's queen attempts to pull off an nseemly power grab from her brother on g2, who suspects nothng. Kasparov calls this "the trng pont of the game, n which both sies have playe brlintly up til now.

 ( ): Two moves occr to us: White can play 26 Bxe5, estroyng a key black

efener of the ark squares, or he can go for 26 f4, issung challenge to Black's queen. Take yor tme here. What woul you play?

26 Rf4? Occam's razor avises that when given a choice, the simplest is usually the best. After White's ncorrect

ecision, opportty vaishes. f you seek to estroy an enemy, then o it without hesitation. n this nstance, the rook takes haf measres, writng his hate sister on g4 a threateng letter nstea, alertng his foe, now on high guar.  Correct was 26 Bxe5! gxh2+ 27 Bg3! (27 Kxh2?? loses to 27 ... Qh5+ 28 Kg2 Qxg6+) 27 ... Qxe4+ 28 Kxh2 Qxg6! (not 28 ... Qx5?? 29 Qh6), when Black stans better with his extra pawn, but White retans some tiative an conversion wil not be so easy. Botv was of the opion that White was n anger of losng, whereas Kasparov, whle partialy agreeng, felt White shoul hol the game. 26  Qh5! A kilng shot. Black threatens h2.

27 Bxe5 Perhaps the best practical chance, snce 27 hxg3? Nx5 leaves White a piece own with his attack rn agron. 27  Qxh2+ 28 Kf3 Qx2 White's queen claspe her pams together n fervent prayer, but they obviously went nanswere. Beko's reang attempts to attack are soon renere useless from his recent rn of reverses. 29 Nf6+ Kg7 30 Nxe8+





The game isn't over yet, espite White's massive aterial eficit. Black must broach the elicate atter of his own kng's safety. Botv allows his opponent some leeway to attack, but oly with the proviso of carefully efne, calculable iits an restrictions. To us humans such positions are stll pretty scary from Black's perspective. Yet Hdini camly assesses it at -7 .74; i.. completely resignable for White! Ufortnately, it isn't possible to supermpose the hun wll upon a position which lacks the essential raw ngreients for success. n this nstance, the camel somehow oes nee fit through the eye of the neele an the rich man is grnte entry through heaven's gates. 30  Kxg6 31 R+ Kh7 32 Bxg3 At last, the checks rn out an now it's Black's trn. White is areay ea. The remaner is simply Black ancng upon the grave. The woul-be attack ens abruptly after 32 f7+ Kg8 33 Rg7+ Kf8. 32  Q3+ 33 Kf2 Sometimes a grievously wone solier feels no pan. f there is pan everywhere n the boy, his bran gets cofuse at the lack of a localze pont of orign, an simply gives the co n: no pan . 33  Qxb5 34 cx6 Qxe8 0-1

Chapter Three Riding the Dynamic Element hen researching this book I was surprised to read Kasparov's statement that Botvik, who we normally associate with ron logic and patient manoeuvrng, was a veritable thaumaturge with the itiative, and worked wonders and mracles when he seized power over the board. In fact, Kasparov claimed Botvik's feel for itiative rivalled or surpassed that of any legendary player n the hstory of the game. As I went through more and more of Botvik's early games, I saw very clearly that Kasparov's assertion was true. n this chapter, we exane Botvk's remarkable hnng of the nitiative, may from his heyay, from the i 1930s to the early 50s. Botvk's isputatious pieces srge forth, always seekng nitiative, always fnng coflict. His nitiative, ike ffile malice, ha a way of growng by feeng on itsef. Even players associate with the nitiative, such as Keres, were often casualy brushe asie by Botvk n his prme. Game  MBo k-MVr Sr Nottngham 1936 Queen's Gambit Declined 1 c4 e6 2 Nf3 5 3 4 N 4 Bg5 Be7

Back n the 1930s, vrtually everyone played the Queen's Gambit Declned response to 1 d4. 5 Nc3 0-0 6 e3 Nbd7 7 Bd3



: Doesn't this lose White a tempo?  'm not a big fan of this move, which obligngly cees a tempo to Black. But it is playe, even toay by top GMs, so it can't be al that ba.  wol go for 7 Rc1, 7 Qc2 or 7 cx5.

7 000 cS  . . . dxc4 8 Bxc4 a6, nducng 9 a4, s more accurate and oly then 9 . . . c 10 0-0 cxd4 11 exd4, when Black reaches a more favourable version of the game, snce he goaded a weakeng of the b4-square. 8 0-0 cxd4 9 exd4 dxc4 10 Bxc4

Tyrants, n order to subjugate, keep the masses n the dark. They bu books,

along with the heretics who read them. To my mnd, Botvk, a similar iron-fisted despot n such structures, was possibly the greatest practitioner of all time of both isolai and hangng pawns positions, nvitng them all his life, especially arisng from hs Nimzo-Indians. If you look at his solaihangng pawn games from the 1920s and 30s, his opponents look like bumblng ncompetents, while Botvik, fused with knowledge which his opponents lack, appears as a modern day GM, like Carlsen or Kramik. Kasparov writes that n such positions Botvik "dsclosed virtually all their resources! He contnues: "But Botvk demonstrated that the activity of the pieces and the pressure n the centre more than compensate for the nsignificant defect n the pawn structure. 10  Nb6 Botvik suggested 1 0 . . . a6 as Black's most accurate move here.

11 Bb3 Bd7 M.Botvik-A.Batuyev, Lengrad 1930, saw 11 ... Nbd5 12 Ne5 Nd7 13 Bxe7 Nxe7 14 Qe2 Nf6 15 Rfd1 b6 16 Rac1 Bb7 17 f3 Rc8? (17 ... Nfd5 was necessary).

 ( ): Black has just blnere. How i Botv psh it?

 Sac on f7 an force Black nto a eath-pn: 18 Nxf7! xf7 19 Qxe6 Qf8 20 Ne4 xc1?! 21 xc1 Nf5 22 N6 Ba8 23 Re g6 24 Nxf7 Qxf7 25 Qxe7! 1-0.

12 Qd3 Nbd5 Black should seek swaps n such isolai positions. Therefore 12 ... Nfd5 may be more accurate.

13 Ne5 Bc6 14 Rad1 : Why i Botvk avoi 14 Nxc6 which picks

Black an isolai on c6?

Up the bishop pair and also hands

 This plan was trie n one game, D.Breer-R.Friman, Geran League 2005. After 14 ... bxc6, Black reforces 5 with a strong grip. This plus the fact that White's e5-kight, a angerous attacker, ay be the superior piece was probably why Botvk rejecte the iea, an  beieve rightly so.

14   Nb4 15 Qh3 Bd5 16 Nxd5 Nbxd5?! The kight moves out of his jursdiction and holds little authority where he stands. This natural yet inaccurate move allows Botvk an attackng build-up on the kngside. Vidmar should have played 16 . . . Nfxd5! . This mnor yet sigificant emendation helps free Black's game.

: But with this recapture oesn't Black also move a efener

away from his kng and

leave his b4-kight danglng on the queenside?

  prefer White after 17 B2 Nc6! (the waywar b4-kight comes back nto play) 18 Bc2 g6 19 Bh6 Re8 20 Qf3 Bf6, but Black's position is not so ba, an certay ftely better than what he got n the game.

17 f4!

From this pont on, Botvik ntersperses direct threats with strengthenng manoeuvres. 17  Rc8 :  realze 17 . . . g6 weakens, but isn't it necessary for Black to halt f4-f5 - ?

 The trouble is that it fals tacticaly to 18 Bh6 Re8 19 Ba4, wng the exchange. Hdini thks the theatic 19 f5! is even stronger.

18 f5! Botvik massages his once rigid structure nto relaxed pliability. 18  exf5? Vidmar grossly underestimates the explosive potential to Wite's game. He had to try 18 ... Qd6.

19 Rxf5 The old black kng's jonts begn to ache from the nclement weather. Wite's rook looms omnously and pressures d5, f6 and f7, all tender ponts n Black's camp. 19  Qd6

19 . . . Rc7 was better, but even then Black is busted after 20 Rdf, and  20 ... Qd6 then 21 Nxf7! Rxf7 22 Bxd5. Clearly, White prepares to make trouble on the kingside, yet the piece destned to perform the dirty deed for now remans shrouded n anonymity. Black just blundered n an already busted position. A hearng s convened and the sentence harsh.

 ( ): How i Botvk exploit Black's last move?

 Deflectioniscovere attack. Force Black nto mtiple, ealy pns.

20 Nxf7! Now White's forces dance with facile ease to the music of Botvik's desres. 20  Rxf7 To negotiate successfully, one must first possess somethng of value the other side desres  a something Black utterly lacks. Vidmar can do nothing but glumly await the further deterioration of his once sound position. 21 Bxf6! ndermng the defender of d5. 21  Bxf6 21 . . . Nxf6 22 Rxf6! exploits Black's danglng rook on c8.

22 Rxd5 Qc6 One wnces at the thought of Black's position. That's a lot of past sn to expiate. Black's queen backs off, exhalng reproachfully, while his kng, precious little life left n him with such grievous threats pendng, now comes to the awful realization that hs so-called protectors are worthless. Some murmur prayers, while others lie around druk.

·



Black's game reeks of unpuished strategic crimes, maly imputed upon multiple underestimations of Wite's power, as his seemngly endless itiative flows unpunctuated and without resstance.

 ( ): How i Botvk fsh the job energeticaly?

 Overloa, snce c8 agan hangs i the offere rook is taken. "This natral abonation is not by Go's esign! rals Black's queen at the offenng rook, who floats to 6 as i propelle by ark magic.

23 Rd6! Note that 23 Rc5?? fails miserably to 23 ... Bxd4+!. 23  Qe8 23 . . . Qxd6 24 Qxc8+ Qf8 25 Qxb7 ends the matter as well.

24 Rd7 1-0 Black's queen and king curse Wite's forces n one language, then plead for mercy n another. f all the games  anotate were this simple, my job woul be so much easier! Viar was a strong GM, yet appeare cruely nept n comparison with Botvk. t felt lke the skil gap wiene as the game went on. Even top GMs of Botv' s ay recognize ther own arke nferiority  which was almost shoute out  n comparison to Botvk n his prme. Such was his omnance from 1936 to the ear ly 1950s. Game  AAlekhe-M Bo k Nottngham 1936 icilian Dence

Botvk acquitte himsef wel n his showown aganst the reignng worl champion, an at the height of Alekhne's powers. Alekhne hmsef wrote: "Botv's wonerful achievement n Nottngham cofrms that he is the most probable caniate for the title of worl champion.

1 e4 cS 2 Nf3 d6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 g6 The obscure (for 1936) Dragon Sicilian apparently didt come as a surprise to the heavily prepared Alekhne, perhaps Botvik's oly equal n the openg phase of the game.

6 Be2

Alekhne would surely have jumped aboard the popular attackng set-up 6 Be3 Bg7 7 f3 0-0 8 Qd2, had it been nvented at the time. 6  Bg7 7 Be3 Nc6 8 Nb3 Be6 9 £4 0-0 10 g4!?

In space, an object may generate ncredible speed n the absence of frictios resistance. Botvk occasionally took on calculated risks; Alekhne, on the other hand, simply loved to gamble. It may be that Alekhne's bran had a curious defect: an elarged lobe which controlled aggression. So to advise him to calm down would be belated and uheeded council. Here we see a brazen attempt by the world champion to put the young upstart n hs place. (nfortunately for Alekhne, Botvik's place at Nottngham was a tie for frst with Capa!) : The risk entaile n White's lnge feels isproportionately bresome to the dreamed of rewards, and it looks borderlne unsound. Is it playable?

 A crme n one society ay be an honorable ee n another.  on't trust it ner the theory: an attack must be comprise of more than just elemental wil; there must exist an nerlyng strategic basis as wel  a basis which  fal to ientfy n this position. But sayng this, beieve it or not, White's stats are qite reasonable after 10 g4!? an it is stil playe by GMs toay, so it must be son or, f faing that, borerne son.

10  d5 Principle Counter in the centre when attacked on the wing. A logical temporary pawn sac. Botvk hopes to eny Alekhne the attack to which he feels entitle. Botv qickly aapts to the rapily alterng circmstances swrng about the centre. Now the combnation of Alekhne's lust for aventre, ngle with Botv's itch for conterplay tears a giant hole n the position's equaiity.

Alternatively, 10 ... Rc8 has scored well for Black;  e. g. 11 f5 Bd7 12 g5 Ne8 13 0-0, M.Bartel-ojtaszek, roclaw 2010, and here I would try 13 ... Bxc3!? (a theoretical novelty) 14 bxc3 Ng7.

: Are yo u serious? Black just gave up his powerfl ark-sqare bishop.

 The reason  suggest givng it up, n orer to aage the opposng structre, is that White's g pawn sits on g, blockng access to h6.  actualy prefer Black's chnces here, but please on't sen me an angry Facebook message f you try my suggestion an get mate!

11 £5 Surgng forward and knockng off a defender of d5.

: sn't 11 e5 better, to keep the centre close?

 t isn't so close after 11 ... 4! 12 Nx4 (12 exf6? Bxf6 favours Black) 12 ... Nx4 13 Bx4, an now n G.Levefish-M.Botvk, Moscow 1936, Black pule an overloa combnation with 13 ... Nxg4! an attane the slightly better position.

11  Bc8 12 exd5 Nb4

13 d6!? Alekhne's novelty.

: Why i Alekhne return the pawn?

 He hope to isrupt the flow of Black's tiative, an there is no way to hang on to the

a) 13 fxg6 hxg6 14 Bf3 has occurred four times and no one found 14 . . . Bxg4 ! (a novelty) 15 Bxg4 Nxg4 16 Qxg4 Nxc2+ 1 7 Kf2 Nxa 18 Rxa Bxc3 19 bxc3 Qxd5. I'm not exactly sure whose kng is n greater danger, but I like Black's chances n ths wild position. b) 13 Bf3 gxf5 14 a3 fxg4 15 Bg2 Na6 16 Qd3, ntendng to castle long next move, as n R.J.Fischer-S.Reshevsky, New YorkLos Angeles (2nd matchgame) 1961, is the usual choice nowadays, when hite looks like he has enough for a pawn. 13  Qxd6 Botvik claimed 13 . . . exd6 was unplayable but Houdini disagrees and offers 14 a3 Re8! 15 Bg5 (certaly not 15 Qd2?? Nxg4! or 15 Bf2? Nxg4! 16 axb4 Nxf2 17 Kxf2 Qh4+ 18 Kg Bxf5 and hite's exposed kng spells big trouble) 15 . . . Nc6 and it's anybody's game. material anyway. For example:

14 Bc5 The bishop looms menacngly, the same way I do when a student dares to yawn loudly durng a chess lesson. The alternative s the crazy lne 14 Qxd6 exd6 15 0-0-0 Re8 16 Bg5 Nxa2+ ! 17 Nxa2 Rxe2 18 Rxd6 Ne8 19 Rd8 h6 20 Kd Re 21 Re Rxe+ 22 Kxe hxg5 23 Rxe8+ Kh7, when Black may be okay snce he uravels with ... b7b6 and ... Bb7. 14  Qf4





The queen contnues to sow agitation. Black can ao try 14 . . . Qxd + 15 Rxd Nc6 16 g5 Nd7 17 f6 exf6! (Botvik suggests the ferior 17 ... Bh8) 18 Bxf8 Nxf8 19 gxf6 Bxf6, when his pawn and bshop pair give him more than enough compensation for the exchange. 15 Rfl! The rook hopes to circumvent the black queen's authority.

: Doesn't Black lack the fns to subsiize his

expensive war? Now hs queen cat

cover the knight on b4.

 This ha been foreseen by Botv. Play on!

15 000 Qxh2 16 Bxb4





If an assasss target s one saturated n power, my advice is: dot miss. The creditors see Black's assets, now n a state of arrears, on the queenside. But fortune is a fickle compaion n times of confusion. Botvk had foreseen this position and had accurately calculated it to a forced draw. The possibilities appear as shiftng shadows of leaves dancng n the sulight.

 ( ): Fn the correct iea an work

out Botvik's sequence. Black to

play and force a draw:

 Sac a secon piece.

16  Nxg4! Through dark powers the shaman, in death, transfers his sprit into the body of the black queen, who rises to take up the battle once again. Botvk actuates the final lunge at hite's king  not enough to kill, but enough to neutralize. ith the sac he tears away the fabric of Alekhne s kng's shelter, now exposed to the elements. The move order 16 ... Qg3+ 17 Rf2 Nxg4! works too. 17 Bxg4 Qg3+ 18 Rf2 The rook block is forced and a perpetual check ensues. 18  Qgl + 19 Rfl Qg3+ 20 Rf2 Qgl + zz This game, lke a suen thnerstorm, erupts, rages for oly a few nutes an, just as suely, subsies. Game 3 M Bo k-S Reshev sky AVRO Tornament, The Netherlans 1938 English Opening

n the mi 1980s  ha the opportty to meet the great Sammy Reshevsky at several Los Angeles tornaments, as well as a smul he gave at the San Diego Chess Club.  got the impression, as  i with Botvk, of a eeply serious personaity, of a person who never cracke a joke or se n his entre fe. n fact, of al the worl class players  met, with the exception of Boris Spassky,  notice they al seeme to share this trait. Or, who knows, maybe  was just ntmiate by beng n their presence, an conjre this mpression up n my own hea.

1 c4 e5 2 Nc3 Nc6 3 g3 g6 4 Bg2 Bg7 5 e3 : Why in't Botvk play his own e2-e4 variation of the English?

 He han't nvente it yet! He worke out his system, orignaly ntene with colors reverse agast Close Siclian, n preparation for his 1954 match with Smyslov. Then it occrre to him: why not play the same system with White, a move up, n the English?

5  d6 6 Nge2

6  Nge7 : Why not play ... f7-f5 an oly then ... Nf6,

rather than post the kight on a passive

square on e7?

 Well, that is exactly what most moern GMs woul play, but that which we take for grante now was smply nown back n 1938. The corse of P.Leko-L. Dng, Beijng (blitz) 2012, 6 . . . f5 7 0-0 Nf6 8 3 0-0 9 b3 Be6 10 N5 Q7 11 Bb2 Rae8 12 Q2, shows a typical set-up for both White an Black.

7 d4 It's a little more common to hold the d-pawn back to d3 n these Englsh lnes.

: But isn't 3 a more passive square?

 Th of it as flexible nstea. By playng the pawn to 3, White retans both futre 3-4 an e3-e4 options.

7 000 exd4!

: Why an exclam to a move which cees White greater central fluence?

 n compensation, Black exerts piece pressre on White's centre an, especialy, a softene up 4-square.

8 exd4 0-0 9 0-0 NfS l0 dS NeS?! Inaccurate. Black should follow the prnciple Seek trades when cramped and play 10 ... Ncd4! 11 Nxd4 Nxd4 12 Be3 Nf, when  piece activity and dark square control make up for his territorial deficit, H.Vonthron-V.Kostic, Viena 2006. 11 b3 as?!





Perhaps the begng of an ncorrect plan. Black ntends . . . Nd7-c, but the trouble is that Wite simply ejects the ntruder with a2-a3 and b3-b4.

: What o you suggest for Black?

 Probably he shoul opt for an ... a7-a6, ... B7, ... Rb8, ... b7-b plan, to generate conterplay.

12 Bb2 Nd7 13 a3! arng Black that he will not tolerate an unvited guest an extended stay on c. 13  Nc5!? Anyway.

14 b4 Nd7 : Din't Black just lose two tempi?

 He i, but he goas White' s pawns forwar n orer to overexten them. Havng playe over hnres an hnres of Botv's games n preparation for this book,  am har presse to fn any where he overextene. Botv's superior strategic grasp gave him the nerstanng of exactly when to push n when to hol back. On 14 . . . axb4 15 axb4 xa 16 Bxa Na6, Botv sai he ntene 17 Ne4! Nxb4 18 g4! with a tremenous tiative for the pawn.

15 Qb3 Botviik would like to enforce an eventual c7-c break. 15  Nd4 Botviik felt ths was an error, but I dot see any great alternatives on Black's part. 16 Nxd4 Bxd4 17 Radl Bg7 18 Rfel axb4 19 axb4 Nf6 20 h3!





ill Neil Armstrong's footprints on the moon still be there a million years from now? Black's counterplay deficit fee like an atmosphereless enviroment, trapped in eternal stasis. It's structive to watch Botvik make progress.

: What is the iea behn his last move?

 White eies his opponent the use of g4, an seconly, he plans to meet ... Bf with g3-g4, anexng even more te rritory.

20  h5 In order to post a bshop on f in comfort.

21 c Gang more space. 21   Bf5 22 Nb5 Intending to harass the bishop with Nd4 next. 22  Bd7?! Almost unconsciously, without volition, Black's forces back up until they reach the lip of the precipice. Reshevsky provokes Botvk's next move, but as the old sayng goes: be careful for what you wish for  you may get it. 23 c6! bxc6 24 dxc6 Bc8





Black's cosmopolitan bshop wallows out of his element n the rural outskirts on c8. This game would be a serious candidate for the Accumulatin g Advanta g es chapter, if not for Botvik's amazngly energetic fish. hite's forces share conspiratorial glances n each other's direction, while Black's ambitions of survival flow n a reverse polarity of decreasng levels of feasibility. Black's once tough defensive barrier now lies fluffy and pliable.

 ( ): Do you see how

Botvik struck a serious blow from this

position?

 Discovere attack ouble attack. The knight offers hsef up as bait to lure Black' s efeners out of ther trenches.

25 Nxd6! Be6 The delnquent bishop can oly shrug and blow out his cheeks n response to hite's open aggression. Reshevsky desperately tries to complicate, seeng that the lne 25 ... cxd6 (the aggrieved party demands heavy recompense, which it will never receive) 26 c7! (dscovered attackdouble attack  Black's queen and rook hang simultaneously) 26 . . . Qxc7 27 Bxa8 Bxh3 28 Bf3 is hopeless for Black. hite simply pushes his b-pawn down the board. 26 Rxe6! Intelligence, cung and resourcefulness are no match when faced aganst a logician, brute-force calculator. Like a savvy nvestor, Botvik has a sharp eye for profitable ventures.  exchange is a trifle to elimnate Black's only active piece. 26  fxe6 27 Nf5!





Discovered attack. The persistent kight nsnuates himself nto Black's busness once agan. Botvik goes after Black's g7-bshop, steward of  dark squares. 27   Qe8 28 Nxg7 After the elimnation of Black's most important defensive piece, the remander of hs tangled forces fnd themselves garlanded with the hangmas noose. 28   Kxg7 29 Rd7+ Rf7 30 Be5!





The bishop's services are freely at anyone's disposal; his blessngs, however, require cold, hard cash up front. The powerful legate on e, already a prnce of the church, brooks even higher aspirations as supreme pontifex of the board. White's rook remans immune and c7 fal, after which White's advanced passers take the day. 30  Kg8 Black's destitute kng lacks even basic ameities and his defenders, completely unequal to the orgaizational task required, scatter n panicked confusion. 30 ... Rc8 31 Qd3 is too horrible to contemplate.

31 Rxc7 The rook's startlng accession to power contnues unabated. 31  Rxc7 32 Bxc7 Ra1 + 33 Kh2 Ra7 34 Be5 Rf7 35 c7 N d7 36 Qc2 Rf8

·



Black's confused defenders stumble about, disoriented. His pieces receive a memorandum from the bank, warng of a lack of funds in his account and imminent foreclosure on all properties. They fnd themselves surrounded by screamng creditors and coivng litigants.

 ( ): Fn one smple move an Black's position crmbles.

 Overloa. 37 c8Q! 1-0 Game  M Bo k-G Lev efsh USSR Championship, Moscow 1940 English Opening 1 c4 e5 2 Nc3 N 3 Nf3 Nc6 4 4!?

4 g3 and 4 e3 are much more commonly played today. 4   exd4 5 Nxd4 Bb4 6 Bg5

: Doesn't Botvk care that Black cn

do great harm to hs structure by taking on

c3?

 Objectively, this lne isn't so hot for White, but Botvk probably felt cofortable n this position, ue to his avocacy of the White sie of the No-nian, where he commoly encourage a bishop to chop on c3. Here atters are more structuraly serious for White, snce he soon takes on a set of ouble isolais on the c-file.

6  h6 7 Bh4 Bxc3+ 8 bxc3 : Those broken pawns are no jokng matter.

What is Wite's motivation behind allowing the damage?

 White gets the bishop par n fair piece activity. oul the game rft towar an enng, then White woul probably regret his ecision.

8  Ne5 I like this move more than allowing 8 ... d6 9 Nxc6 bxc6 10 c!, Smirn-S.Tiviakov, Rostov on Don 1993.

9 e3 After 9 f4 Ng6 10 Bxf6 Qxf6 11 g3 Nf8 12 Bg2 Ne6 13 0-0 0-0 14 e4 d6 15 Qd2 Nc 16 Rae1, Wite's greater central nfluence compensated for his nferior structure, V.Kramnik-A.Karpov, Las Palmas 1996. 9  Ng6 So as to break White's pn without weakenng with ... g7-g. 10 Bg3 Ne4!? Often, that which we want and that which we actually need appear contradictory. Levenfish elimnates Botvnnk's bshop pair, when it may actually be better to refran from this move and simply develop with 10 ... d6.

: Why?

 Hanng White an open h-file mens that Black's kng never rests easy castlng kngsie. Queensie castlng remans tricky ue to the open b-file.

11 Qc2 Nxg3 12 xg3

·



Wite's extra space and attackng chances easily compensate for his damaged cpawns. 12  d6 13 f4! Grabbng more space and denyng Black use of e5. Already we sense hostile overtones from Botvik's side.

: But in't White just fict a backwar e-pawn upon hmsef?

 The backwarness of the pawn is merely cosmetic, snce Black is n no position to a pressure to it.

13  Qe7 14 Kf2 The correct home for the kng, where it is quite secure and helps out defendng e3. 14  Nf?! After t move Black's mighty labours are rewarded with distressngly dimnutive returns. The ntention  probably to head for c, but Black doest have time n an open position for such retrograde luxuries.

: What o you suggest?

  lke my Everymn cousn M ichar Paliser' s treatment: 14 . . . c! (Black grabs much neee central fluence) 15 Nf3 (perhaps White shoul try 15 Nf5, but he probably feare too many swaps with his ferior structure  Black ay contnue 15 . . . Bxf5 16 Qxf5 Qe6! to alow queensie castlng) 15 . . . B 7 16 B3 --! an suely  prefer Black's position, snce his kng looks safe enough n he retans structural benefits, V.Krutti-R.Paliser, York 2000.

15 cS!

·



Botvik's imble mind rapidly adjusts to the changed circumstances, and from this moment, he seizes the itiative and never lets go. 15  dxcS?  apology to the deity before commission of the crime hardly counts as an expiated sn. ith his last move, Levenfish impregnates hs position with more optimism than efficiency. He should balk at an offer of early confrontation when he clearly ist ready for it. nfortunately, we all desire that which we don't possess. Hey VIII romanced Mary Boleyn, when all along he desired sister Ane. How to reconcile the discrepancy of yearng with the harsh vicssitudes of ill fortune? Black soon regrets hs decsion and the vagaries of a fickle wnd blow his position far off course. retched as it looks, Black had to try 15 ... Nh7! (headng for e4 or g4) 16 cxd6 cxd6 17 Bb5+ Kf8 with an nferior but still playable game. 16 BbS+! Nd7? Black has to try 16 ... Bd7 (16 ... c6?? loses to 17 Nxc6) 17 Nf5 Qf6 18 Qe4+ Ne6 19 Bxd7+ Kxd7 20 Rhd+ Kc8 21 Rab Rb8, when he still hangs on, albeit just barely.

17 Nf5 Qf6 The queen rases an eyebrow, with ample justification.

18 Rad1 Just like that, Black is helpless. 18  g6 18 . . . a6?? loses immediately to 19 Qe4+ Kd8 20 Bxd7 Bxd7 21 Qxb7 Rc8 22 Rxd7+ ! and mates. 19 Nxh6 Rf8 20 g4! The kngside brims with hite's hostile ntent. 20  a6 21 gS Qe6

 (  ): What is the best post for

hite's bshop? Come up with an attackng

plan.

 That which is beneficial to white's bishop is also goo for his brother on h6, n vice versa. The bishop convenes a clanestne meetng with his h6 conterpart by takng control of g4, fter which White threatens to swng his knight ron to either to e or f6, whle clearng the path for 7.

22 Be2! Nb6 : Why i Black move his efener of f6

away and allow hite a devastatng kight entry?

  true, but what else can Black try? He is n virtual zugzwang. For example, 22 ... a 23 Ng4 an then what? Black is paralyse.

23 Ng4 Ke7 24 Nf6 Black s unable to reman outside the breadth of the toxic kight's fluence. 24  Qc6 25 Rh7 reateng Qxg6. 25   B£5 26 e4 Be6 27 £5 1-0 Oh, the wondrous possibilities. Botvk looks here, looks there, like a bedazzled eight-year-old at the county fair, not knowng which ride or what food to begn with. Black's game morphed nto an allegory on the evils of sloth.

:  realze Black is n ba shape, but is this position really resignable?

 Houdini assessment: + 6.1 5( !)  more thn a roo k up i converte nto the realm of the material! n every variation Black simply gets pushe off the boar.

One possible ightmarish future for Black: 27 f5 gxf5 28 exf5 Bd7 29 Bf3 Qa4 30 Qe2+ (the ruthless quees eyes gaze coldly on those foolish enough to cross her) 30 . . . Kd8 (the exhausted black king slumps over, limp as an old, worn-out pillow; he once couldn't conceive of a world without himse in it, and now must reconfigure his thoughts, and contemplate his own imminent non-existence) 31 Rxf7! and  31 ... Rxf7 then 32 Qe8 is a brutal mate.





A complete ahilation of one of Botvk's key Russian rivals of the 1930s. Game 5 PKereMBotvk USSR Absolute Championship, LengraMoscow 1941 Nimzo-Indian Defence 1 4 N 2 c4 e6 3 Nc3 B 4 Qc2

Capablanca's variation. 4  d5

4 ... 0-0 and 4 ... c are Black's main lnes.

5 cxd5  a3 is also played here. S 000 exdS





e reach a position akin to the Ragozin Variation of the Quees Gambit Declned, one of Black's sharpest options, where he often sacrifices both structure and bishop pair to gan a lead n development and early itiative  the oly difference beng that hite has played Qc2 here, nstead of Nf3. Botvik was fond of  ... Qxd throughout the 1930s. For example: 6 Nf3 (or 6 e3 c 7 a3 Bxc3+ 8 bxc3 Nbd7 9 Nf3 b6 10 c4 Qd6 11 Bb2 Bb7 and Black has developed harmoiously and looks okay, despite hite's bishop pair, M.Euwe-M.Botvik, Nottngham 1936) 6 ... c 7 Bd2 Bxc3 8 Bxc3 cxd4 9 Nxd4 e 10 Nf3 Nc6 11 Rd1 Qc, and Black's superior development and greater central nfluence compensate for hite's bishop pair, G.Levenfish-M.Botvik, MoscowLengrad (7th matchgame) 1937.

6 Bg5 6 7 B4 c : t looks to me ike Black tries to wrest

the itiative n an unjustified maner. Am

I correct?

 t has always looke that way to me too, but the variation is son an White must know what he is ong or face an early loss of tiative, as occrs n this game.

 O-O-O?

·



The devl nvariably offers uimited power. Just make certan you read the fne prnt before signg any contracts. Wite's kng suffers from a congenital impairment which leaves him no match for his healthy twn on e8. sually I don't dole out question marks early on n older games, but ths tme Keres had it comng! This move is weak, even n the context of 1941. Hs theory of parallel evolution n separate solar systems doest apply n this nstance to hs own attack, which fai to materialize n ths game. Hs somewhat outlandish decision s based upon the philosophy: that which s never strived for, never transpires.

:  fal to see the problem. What is so awful

about Keres' idea of itiatng opposite

wng attacks?

 f two prnces of the ream vie for the throne by engagng n a civl war, then you ha better not pick the pretener's sie as an ally. Here we have a situation where Black is clearly faster, ue to the soon-to-be open c-file, where White's kng an queen hule aganst the comng ... Rc.

: f the move is as ba as you claim,

then what was Keres' motivation behnd it?

  conjectre: Keres nerstoo that he was clearly Botv's ferior strategicaly an may have felt that his best shot was to n thngs up. The problem was that Keres (an many others, nclung mysef n my smul game aganst Botv!) nerestiate Botv's power tactically an with the itiative, which was at least equal to Keres level, i not superio r. Seconly,  beieve Keres navely  lke a person who tells his nterviewer: "1 wol ike the job, but  have absolutely no aptitue or work experience with it!   hope to get the better game by followng n the footsteps of Mkenas-Botv playe a year previously (see the next note).

But here Keres learned a panful lesson: never ever repeat a lne aganst Botvik, uless you are ready to face the wrath of hs formidable home preparation. 8 dxc5, 8 Nf3 and 8 e3 are all superior options for Wite.  000 Bxc3! Superior to 8 ... 0-0 9 dxc5 Bxc3 10 Qxc3 g5 11 Bg3 Ne4 12 Qa3 Be6 13 f3 Nxg3 14 hxg3 Qf6 15 e3 Rc8 16 Kb1, when Wite stood a shade better n V.Mkenas M.Botvk, SSR Championship, Moscow 1940. 9 Qxc3 g5! This knd of move is routne today, though it was considered a radical weakeng back then.

: Why oes Black get away with such a blatant weakeng move?

 For one smple reason: he owns, an wil never relnquish, the tiative. White simply isn't given the tme to exploit Black's structural weaknesses.

10 Bg3 The fuming bishop clearly deems Black's last move an impertinence. 10 000 cxd4! A theoretical novelty for the time. Black takes on d4 before hite can play Nf3 or e2-e3 and recapture on d4 with knight or pawn. Botvik's move is a clear improvement over 10 ... Ne4, as played in 5 Belavenets-Vimagin, Moscow 1941. 11 Qxd4 Nc6 12 Qa4 Bf5!

Dual purpose: 1. Cutting off a b1 escape route for hite's king, who now lounges in the centre. 2. Black clears c8 for a rook.

13 e3 If Botvik had a character flaw, it was that he loved to trash hs riva n his anotations: "The deficiencies of Keres' character tell. He was unable to endure the stoically unpleasant surprise, and he misses an obligatory opportunity to complicate the play: 13 f3 Qb6 14 e4 dxe4 15 Kb1, movng his kng away from the threats of the enemy pieces. However, n Kasparov's opion, ths position is completely hopeless for hite. Houdini backs him up assessng it at a dsmal -5.36! So perhaps Botvik owes Keres (and Fischer too!) an apology! 13 000 Rc 14 Bd3?! A ha-fuilled promise s a half-broken one as well. Keres hopes to cleanse his kngdom of an evil nfluence but oly manages to make matters worse. Now hite's h-rook and g-kight clearly suffer an aversion to hard work and are content to reman n stasis, unpromoted. Then agan hite's game was beyond savng, even after the superior 14 Ne2 0-0 15 Nc3 Ne4 16 Nxe4 Bxe4, and now hite must play the god-awful 17 Kd2, a move no chess player wants to make when facng the strongest player n the world across the board! 14 000 Qd7!

Black breaks the pn on c6 and threatens . . . Nb4+ . Both queens arrive at the party wearng the same blue dress  a dress which both believe fai to suit the other n the least. 15 Kbl Bxd3+ 16 Rxd3 This guy lumbers up the board with cumbersome, mastodoic apathy. 16  Qf5!





Botvik embroiders a not-so-delicate attackng pattern. Now the ncongruous parts converge nto a harmoious whole. e see the second deadly pont behind 14 ... Qd7!. hite s forced nto a death pn.

17 e4 No choice, snce 17 Qb3 Nb4 s crushing. 17  Nxe4 18 Kal The white kng's haggard face feels dsproportionate to his actual years. 18  0-0 Threateng ... Nc5.

19 Rd1 : Why not 19 3  ?

 White's back rak is vunerable an Black respons with the crushng shot 19 . . . N4! , wng on the spot.

19  b5! The general's philosophy: kill first; negotiate later.

20 Qxb5





Wite's queen rases eyebrows in sharp, silent rebuke, oly to be hit with a devastatng counter next move when she turns her back.

 ( ): Fn a path which fshes White's resistance.

 Back rnkfork overloa. Black's kight reaches out an smites, ike the Ol Testament han of Go to the isobeient an the wicke.

20   Nd4! 21 Qd3 Nc2+ 22 Kb1 A prisoner harbours two primal thoughts: 1. How to avoid dscomfort. 2. How to escape. In this nstance, Keres' kng achieves neither. 22  Nb4! 0-1





If Wite's queen moves, Black uleashes a horrific dscovered check. How oes one go about efeatng a leang contener for the worl chess crown n 22 moves an with the black pieces? Somehow, Botvk mnage to pul it off, akng Keres look ike a bumbng aater n the process. Botvk couln't resist statng the obvios: "White's kngsie pieces took no part n the game. 'm pretty sre Keres was wel aware of this happy fact! Game  IBoavsky-MBotvk

USSR Absolute Championship, LenngraMoscow 1941 rench Defence 1 e4 e6 2 4 5 3 e5 c5 4 Nf3

: Why in't White back up his centre with 4 c3 - ?

 Bonarevsky tries a Niowitsch iea, where he temporarily sacs his 4-pawn. Later on White plays Nb2 an Nb3, hopng to regan his pawn with a blockang piece on 4.  thnk it's a sub-par lne because Black eventualy eroes away White' s centre entrely  although, havng looke at the atabase stats, White seems to o okay with it,  suspect, ue to its srprise value.

4  Nc6 5 Bd3 5 c3 transposes to the man lne.

5  cxd4 : Why not gan a tempo with 5 . . . c4 - ?

 Because Black gives up too much for a mere tempo. By playng . . . c5-c4, Black releases the tension on 4, alowng White's centre to r eman nchalenge  after which White has the leisre t o use his space an central nfluence to bul up an attack.

6 0-0 Bc5 7 a3 If Black underestimates the danger hesheit can easily walk nto dsaster, as n this game: 7 Re Nge7 8 Nbd2 -?? 9 Bxh7+! (the bshop piously rases his eyes heavenward and sacrifices himself on h7) 9 . . . Kxh7 10 Ng5+ Kg6 11 Qg4 and hite had a wg attack, R.Dzndzichashvili-Comp Fritz, New York (blitz match) 1991. 7   Nge7 8 Nbd2 Ng6 Havng undermned the e5-paws support, Black now pressures the head of the chan.

9 Nb3 M.Crosa Coll-Lui.Gonzalez, Mendoza 2004, saw 9 Re f6 10 Nb3 Bb6 11 Bxg6+ hxg6 12 Qd3 Kf7 13 Bf4 g5 14 Bg3, when I have grave doubts about hite's

compesatio after 14 ... g4 15 Nfxd4 Nxe5 16 Qe2 (16 Bxe5? fxe5 17 Rxe5 Bc7  eve worse) 16 . . . Bc7 17 Radl.

: sn't 9 b4 White's most logical move here?

 Yor move sezes queensie space with tempo an leaves open pos siblities of Bb2 i necessary. Criously,  on't see a sngle game with this move n the atabase. So perhaps the problem ies n the lne 9 ... Bb6 10 Re Qc7, after which White has trouble efenng his e-pawn. He can either sac it, which looks fishy, or he can play 11 Bxg6 hxg6 1 2 Nb3, when he reaches a position siar to what Bonarevsky get s n the game, except that it isn't clear i the nclusion of b2-b4 actually helps White.

9 000 Bb6 10 ReI B d7 11 g3

: What is the pont of this move?

 Dual prpose:

1. White iteds h2-h4-h5, chasig away the g6-kight, but Botvk simpy doest give Bodarevsky the time. 2. White prevets a future . . . Nh4, or eve . . . Qh4, tricks. Istead, L.Matsev-J.Bai, Word Juior Champioships, Kemer 2009, saw 11 Bxg6 hxg6 12 Bf4 (12 Nbxd4?? Nxd4 13 Nxd4 Qh4 ws, due to simutaeous attacks o h2 ad d4  ad f2, shoud White retreat his kight to f3) 12 . . . Rc8 13 Qd3 a! 14 Racl a4, ad ow White fe for the same trap with 15 Nbxd4?? Nxd4 16 Nxd4 Qh4! ad was busted, havg too may oose pots   positio. 11 000 £6 A direct chaege to the head of the cha. 12 Bxg6+ 12 e xf6? ! Qxf6 woud simpy give Back a automatic attack dow the f-fie. 12 000 hxg6 13 Qd3 : Why oes White settle for a trncate version of his orignal

desire whe he ca

simpy rega the ost paw with 13 Nbxd4 - ?

 This may be White's best bet, though  stll ike Black with his bishop par an open h-fle.

13  Kf7 14 h4

 (  ): White plans Bf4 an then the leisrely recaptre of his lost

d4-paw. How did

Botvik competey cross him up ad disrupt this pa?

Warg: T oe is ubeievaby difficut to sove. If you fid t idea you are a GM or wi be oe soo!  The sorceress traces strange patterns with her forefnger n the ar, which begn to swrl with power. The queen, gracefl an supple, glies about on the contors of the back rak to enter the attack. Botvik embarks upon a quxotic  one may argue reasonable  iea. His plan, opaque as froste glass, comes across as artficial, yet it works.

14  Qg8!!

: What on earth?

 A move which leaves an neible imprnt upon or mns.  realze the move looks ncomprehensible, esigne to transport who-knows-what-piece-to-who-knows-where? The iea is actually quite straightforwar: . . . Qh7, folowe by . . . g6-g5! which pries open White's kng, even though it may soon be an enng.

15 Bd2 : Why i White back off from his Bf4 plan?

 f 15 Bf4 Qh7 16 Nbx4?, then Black strkes with 16 . . . Nx4 17 Nx4 g5! 18 Qxh7 xh7 (two white pieces hang) 19 Be3 gxh4 an White rops a pawn.

15  Qh7 15 . . . Rh5! is aso powerfu, creasg pressure to e5 ad thikg about addg major pieces behid the rook ad the foowg with .. . g6-g5.

16 Bb4 Itedg to eter to d6.

16  g5!





The impuse to sac bubbes up. Now Botvik's forces erupt to frezied activity aroud Back's kg, eve  the absece of quees o the board.

17 Qx7 Rx7 18 exf6 If 18 hxg5 fxe5 19 Nxe5+? the 19 ... Nxe5 20 Rxe5 Rah8 21 Kg2 d3! (threateg ... Rh2+) 22 Kf3 dxc2 23 Rcl Ba4 ad ite is busted. 18   gxf6 19 hxg5 e5! Just ook at Back's rog cetre!

20 gxf6 Kxf6

: Does Black have enough compensation?

1. A ope h-fie ad serious threats to hite's kg. I fact, he eve has access to a ope g-fie, which he ater uses to advatage as we.  For the pawn, Black receive:

2. A powerfu, cetre, just itchg to surge . 3. The bshop pair  a ope positio. 4. A exposio of ight square activity from hs bshop, ow uchaed ad uopposed. 5. Back's kg, uike his cowerg couterpart, participates  the fight, boterg the cetre. Ths meas that Back essetiay has a bous, three extra fightg uits o hs side. Cocusio: Dscoutg a mirace, the odds of White's surviva are egigibe.

21 Bd6 ReS Cautious, but there is o eed for it. He shoud pough ahead with 21 ... e4! 22 Nh4 Rg8 23 Kfl Ne5! 24 Bxe5+ Kxe5 25 Nf3+ Kd6 26 Nfxd4 Bg4! 27 Kg2 Rgh8, whe White is hepess to dea with the comg rook vasio, sce 28 Rh1 is met by 28 . . . Bf3+, wg o the spot.

22 N4 This oses  as do a other tries. 22  Rg8 Threateg a cheapo o h4.

23 K2 The kg hides  the corer, his face a ambet picture of terror. After 23 Nf3 d3! Back threates both c2 ad . . . Rxg3+ . 23  Bf5!





Aother attacker brought to the mix. Now ... d4-d3! becomes a major threat.

24 Re2 Coverg c2. The rook makes a impassioed pea for cam ad co-operatio to hs paic-stricke brethre, who refuse either to iste, stay cam or co-operate. He soo ears a harsh esso: a sge dissetg voice is o match for a authoritaria rege. 24  d3! 25 Rd2 The equivaet of resigatio, though 25 cxd3 Bxd3 26 Rd2 Bc4 27 Nc1 (27 Nc5

Rd8 ws) 27 ... Nd4 is crushg ayway. 25 000 dxc2 26 f4 26 1 Ke6 27 Bc5 Bd8! ws. Bodarevsky makes a desperate attempt to mix it up, but fai to cofuse his oppoet. 26 000 Be3! Yet aother member of the attack force begs to haut the vicity of hite's kg. 27 Bxe5+ Nxe5 28 fxe5+ Ke7! 29 Rf1 Istead, 29 Rg2 is met by 29 . . . Be4, whie 29 Rxd5 Rxh4+ ! 30 gxh4 Bf4+ 31 1 Be4 is mate.





The rug authority coapses ad the makeshift terim govermet is ceary adequate to dea with the baefu threats. Bodarevsky hods o by the most teuous wsp of threads, but ow it fay saps.

 ( ): Black can simply capture the rook on 2, but

there s a more deady

ad more eegat fish  the positio. How ca we expedite the process ad force ite's immediate resigatio?  29  Q!  Overloa eflection. To play on constitutes an exercise n futlity: 30 xc1 (the absence of White' s £1rook allows a matng combnation; 30 Nxc1 Bx2 31 S Bxc2 wns a whole rook) 30 ... xh4+! (forensic analysis of bloo spatter patterns nicate that the kiler is left-hane) 31 gxh4 Bf4+ 32 Kl Be4+ an White's kng stares with ncreuity at the nsolence of the upstart on e4. Soon, he breathes his last, his face set n the severity of the ecease. Game 7 MBok-MEuwe Worl Championship Tournament, The HagueMoscow 1948 emi-lav Defence  4 5 2 Nf3 Nf6 3 c4 e6 4 Nc3 c6 5 e3 Nb7 6 B3 xc4 7 Bxc4 b5 8 B3 a6

hat oce was ew s ow od. The combatats e for cotro  what  currety a fashioabe e of the Semi-Sav.

9 e4 c5 10 e5 10 d5 is ite's major optio  the positio. 10   cxd4 11 Nxb5 The kight se himse as deary as possibe before ite captures f6. 11  axb5 Cosidered superior to 11 ... Nxe5 12 Nxe5 axb5 13 Bxb5+ Bd7 14 Nxd7 Qa5+ 15 Bd2 Qxb5 16 Nxf8  hite teds to score we  this positio after either recapture o f8.

12 exf6 Qb6 Botvik ad Euwe, two of the giats of opeg theory of their time, agree to eter the widest recesses of a aready bewiderg variatio, igorg its terrors or privatios. The moder move order s 12 . . . gxf6  it s geeray beieved that this positio cost Kramik hs word champio tite. He suffered two stgg osses, may due to ad's superior opeg prep,  their 2008 match. Oe caot ose two games with the white pieces ad the savage a tite match  13 0-0 Qb6 14 Qe2 Bb7!? 15 Bxb5.

: s the sac of b5 son?

 Opnions natraly ffer wiely, even among very strong players. My feeng is absolute, ncontrovertible evience isn't always necessary when ecing to embark on such a corse. t just looks right. For example: a) 15 ... Bd6 16 Rdl Rg8 17 g3 Rg4!? with dizzyg compicatios.

V. Kraik-V. Aad, Word Champioship (3rd matchgame), Bo 2008. b) 15 ... Rg8 16 Bf4 Bd6 17 Bg3 f5 18 Rfcl f4 19 Bh4 Be7 20 a4 Bh4 21 Nh4 Ke7 ad your guess is as good as me here, V.Kraik-V.Aad, Word Champioship (5th matchgame), Bo 2008.

13 fxg7 : This helps Black evelop. Can White sac

to crease his ead  deveopmet with

13 0-0 - ?

 That is quite possible, when play woul ikely transpose to the Krak-Anan games above after 13 ... gxf6 14 Qe2.

13 000 Bxg7 14 0-0

14  Nc5!? Risky, but probaby payabe if Back foows through correcty.

: Why? t is playe n accorance to prnciple:

Bishops are more valuable than kni g hts in

an open position.

 True enough, yet Black goes fter the bishop par at the cost of tme, n violation of the spirit of the precept he ntene to follow. From my experience, a lea n evelopment means more than bishop pair n open games. Moreover, Black relnquishes control over the key e5-square, a factor Botvn was later able to exploit. 14 . . . Bb7 an 14 . . . 0-0 are common an, n my opnion, superior alternatives.

15 Bf4 15 Re may be sighty more accurate, retag optios for hite's dark-squared bshop. 15  Bb7 16 Re Rd8?! is move ooks okay, but be wared: the stats show a 9% success rate for Back!

: What  o yo u suggest nstea?

  ike Black's treatment n the followng game: 16 ... Nx3! 17 Qx3 Bxf3! (relnquishng the bishop par but, n ong so, lutng White's kngsie attackng chances) 18 Qxf3 0-0 19 Qg3 8, when Black has a central pawn majority an a safe (for now) kng, E.Bogoljubow-P.F.Schit, Salzburg 1943.

17 R1 Rd5

·



Euwe fights for cotro over e5, but Botvik sti ows the square.

 (  ): Come up with a plan to strengthen White's position.

 Einate Black's best piece.

18 Be5! Bxe5? After this move Back's positio degeerates quicky. Houdini suggests the rather desperate-ookig exchage sac  . . . Rg ! ! (a aomaous move which oy a computer coud fd) 19 Bxh7 Bxe5! 20 Bxg Bf4!, after which it ooks ike he gets reasoabe pay for the exchage. 19 Rxe5 Rxe5?!

: Why i Euwe allow the white knight a free jump nto e5?

 t oesn't look right to me either. Euwe probably in't want to rop a pawn n the lne 19 ... Ke7 20 x5 Bx5 21 Nx4, but at least Black reans active after 21 ... Rg8.

20 Nxe5 reateg Qh5. 20  Nxd3 Back, fearg for his kg's safety, uderstadaby fee a urgecy to swap dow. Istead: a) 20 ... Rg 21 Bf1! (threateg both the d4-paw ad Qh5) 21 ... Nd7 22 Qh5 Nxe5 23 Qxe5 Ba6 24 a4! is hopeess for Back as we. b) 20 ... h5 21 b4! Nxd3 22 Qxd3 Rg 23 Qh7! ad if 23 ... Rxg2+? (a uwise cursio sce White's kg is ampy fortified with abe defeders, whereas Back's ist so ucky) 24 l, whe there is o good way to cover f7.

21 Qxd3 £6

·



Eve i the most terrifyig atura dsasters, there s aways a hadfu of recacitrat citizery who refuse to eave ther homes. Back's kig s dested to rema  the cetre.

 ( ): One glance tels us that Black's

kg experieces serious

difficuties. How did Botvik maage to get at it?

 Sac the kight n orer to fltrate at g7. The black kng nervously mops his forehea with a hankerchief upon hearng that rumors begn to fly an prolferate about his nabiity to rule the kngom. Meanwhile, hite' s queen brushes asie al petty, nterveng fnctionaries an eans a private auience with Black's kng.

22 Qg3!! ite's kight, to the exasperatio of Euwe rejects the phiosophy: after reachg the brk of a precipice, there is o other sae drectio but reverse. Ad so he remas trasiget  his cotued refusa to remove himsef from e5. 22  fxe5 ite's kight may be goe, yet somehow its presece is sti fet, ike the spectra death-shriek of a og-dead warrior. Back may as we accept, sce if he refras from capture, Back tois without pay. I the year 1456, astroomers spotted what is ow kow as Haey's comet, see streakg across the heaves. I those days the commo beief was that comets were portets of i tidgs ad the precursor to the fa of etre atios. So Pope Caixtus III saved the day. Hs soutio: excommuicate the comet! I'm guessg Caixtus' actio didt bother the comet much, just as Euwe's choppg of the e5-kight didt exacty worry Botvik. 23 Qg7 Rf 24 Rc7! The pot. Back must had over his quee to avoid immediate aiatio. 24  Qxc7 The disorieted quee sees oy burred shapes ad muted coours dacg before her eyes.

25 Qxc7 Tears we up  the back kg's swoe eyes at the sight of  ow deceased quee. 25   Bd5 26 Qxe5 d3

The d-paw  Euwe's fa prayer. He hopes to odge it deepy eough to eemy territory to force ite to take perpetua check. 27 Qe3 Bc4 28 b3 Rf7!

: Why i Black han over a piece?

 Euwe contnues to offer tenacious resistnce n the face of nsrmontable fficulties. 28 ... B5 29 Qx3 is resignable for Black. The eeply entrenche -pawn is Black's lone hope. When heavly own n material, the losng sie has ittle to lose an much to gan n contnung the generosity to fee an iea, even one conceive n esperation's womb. Nevertheless, Euwe's clever iea is flawe by a sngle nsrmontable obstacle: it oesn't work f White follows through correctly.

29 £3 ite's kg heads for d2, the key bockade square. 29 000 Rd7 30 Qd2 Most certay ot 30 bxc4?? d2, whe ite must take perpetua check with 31 Qxe6+. 30 000 e5 31 bxc4! Now he ca take it. 31 000 bxc4 Back threates to defect the quee with  c-paw ext.

32 K£2

·



The reforcemet arrives, as coo rai to a witig, thirsty pat. At first gace it appears as if Back's passers cotue to hod the white kg ad quee hostage, but Botvik easiy sees through the attempted subterfuge. 32  Kf7 : Can't Black raw with the eflection sac 32 . . . c3 33 Qxc3 2 - ? White's queen cn't get back to , so he must now take a perpetual check, correct?

 ncorrect. White simply returns the queen to wn a trivialy easy kng n pawn enng with 34 Qc8+ Ke7 35 Qx7+! Kx7 36 Ke2, elmnatng the renegae -pawn.

33 Ke3 Ke6

 (  ): How can White transform his

cosiderabe materia advatage to the

fu pot?

 Transfer the kng to the job of menial labour on 2 to free White's queen.

34 Qb4! A move which effectivey ocks the gate o Back's further cheapo attempts. 34  Rc7 34 . . . d2 fais to the simpe 35 Qxd2.

35 Kd2 Back's passers have bee costraied, eavig hite's a-paw free to surge to its queeg square. 35  Rc6 36 a4 1-0 Game 8 VSmyslov-MBotvk Trang match, Moscow 1952 Dutch Dence 1 4 f5

Botvik ormay bega his Dutch es with the 1 . . . e 6 2 c4 f5 move order.

2 e4 The Stauto Gambit, greaty feared at cub eve but reay ot so hot for hite, accordg to curret theory. I reaity, the e, much ike a disparagg commet, oy has the power to det Back's ego, rather tha fict physica pa. Sometimes payers who essay this shaky gambit emerge with a scap, but ofte tha ot, the scap eds up beg from their ow heads.

: Why is that?

 n the age of computers, most gambits wnle an ie. The comps are just too savvy at grabbng an holng the extra material.  we have to o as humans is to spar with them, an watch an learn their techniques.

2   fxe4 3 Nc3 Nf6 4 g4!?





Reaso ad ogic coapse whe attemptg to egotiate with a upredictabe ad ureasoabe foe. Payed uder the theory: sometimes the oy way for a esso to sk  s to teach with a pair of swgg fsts.

4 000 h6!

: Doesn't this move ramatically

weake the ight squares aroud Back's kig?

 t's a calculate risk, an probably a goo one. His efensive armour may not look pretty, yet it proves capable of withstanng a fearsome blow without piercng.

Back oses a ot of time  he refuses to weake. For exampe, after 4 ... d5 5 g5 Ng8 6 f3!, White gets a especiay mea-spirited-ookg versio of the Backmar Diemer, as Back ags dagerousy behid  deveopmet, S.Tartakower-J.Mieses, Bade-Bade 1925. S Nh3!? A ew move  the positio which ooks ike a poor oe, ad has ever bee  tried aga, probaby due to this game. Smysov gathers the data but coates to a fauty ferece. Hs move has the effect of aduteratg a aready wag itiative. Soo, Smysov ears:  you risk cofrotatio, you must ao be psychoogicay prepared for a rebuff.

: How shoul White contnue?

 He shoul chip away at e4, startng with 5 f3, though even then,  on't believe n White's compensation fter 5 ... 5 6 h3 Nc6, D.Bronsten-M.Gurevich, Moscow 1987.

S 000 dS 6 f3 cS!?





Botvik ups the ate ad compicates further, attemptg to dismate White's cetre competey. I woud go for the more thematic 6 ... Nc6, simpy because it is a deveopg move. Back teds ... e7-e5, freeg hs kgside pieces.

7 dxc5 7 g5 hxg5 8 Nxg5 Nc6 9 fxe4 cxd4 ao ooks rather sicky for White. 7 000 eS 8 fxe4 Bxg4 The simpe 8 ... d4 ooks quite promsg for Back too.

9 Qd3

Back takes cear cotro after this move. Perhaps it was time to go psycho with a specuative piece sac, startig 9 Be2!? Bxh3 10 Bh5+ Kd7 11 exd5 (threateg Qf3) 11 . . . Kc8 12 Be 3. hite may have some compesatio here with those timidatg cetra paws ad a disrupted back kg. Sti, a piece is a piece. 9 000 Bxc5! Moses woud agree: "A eye for a eye . A tooth for a tooth. Botvik refuses to get outsac' ed by Smysov ad keeps pace by prompty retug hs extra materia to crease his growg ead  deveopmet. Persuasio or coercive threats had itte effect o Botvk who, from the very start, desred oe thig oy: the itiative. The archbshop o c dreams of oe day becomg a carda, just as the carda o g4 dreams of becomg pope. Our good ad evi are sometimes compartmetaized: kights of the midde ages coducted themseves with a strict chivaric code of hoour, yet at the same time, they committed savage, murderous acts upo their eemies o the fied of batte, bured crops ad hoy books, piaged ad murdered peasats who had the msfortue to ive uder the eemy's doma. I the same way, Botvik goes to ki mode, ayg waste to a  his path. He seeks itiative ot materia, ad ist sidetracked with bribery es ike 9 ... d4 10 Nd5 Nxd5 11 exd5 Qxd5 12 Qg6+ Kd8 13 Qxg4! Qxh1, whe hite's ight square grip, couped with Back's secure kg, offers some compesatio for the exchage.

10 exd5 0-0

: How woul you assess this position?

 Avantage Botv. Material is even but Black leas i evelopment i a wie open position.

11 Be3 The piece sac 11 Bxh6 appears refuted by 11 ... gxh6 12 Qg6+ Kh8 13 Qxh6+ Nh7 14 Bd3 Rf7 ad if 15 Rfl Rxfl + ad  16 Kxfl Qf6+, whe quees come off the board. 11 000 e4! The e-paw, which caot be captured, aows Back a fitratio hook to f3 ad acts as a caustic aget, sowy erodg hite's patiece.

12 Qd2 Not 12 Nxe4?? Nxe4 13 Qxe4 Bf5 ad hite ca resig. 12   Bxe3 13 Qxe3 Bf3 14 Rgl Ng4! 15 Rxg4 There is o egthy mourg period for a pague victim. The rook's corpse gets tossed uceremoiousy ito a ime-soaked ditch. No choice ayway, sce ... Qh4+ was  the air. 15   Bxg4 16 Nf2 Bf3 17 Bh3!? After recet depetios, hite tries to make do with a Sparta attackg force, but there just s o attack to be foud. Istead, Smysov shoud probaby see to the wefare ad comfort of his kg, startg with 17 Kd2. 17  Qd6 18 Be6+ Kh8 19 Kd2 Na6 20 Re Nc7 Ejectg the truder. 21 Nfxe4 Qxh2+ 22 Kcl Rae8 23 Qc5 Qf4+ Or 23 ... Bxe4! 24 Nxe4 Qh4! (doube attack) 25 Re2 Rf1+ 26 Kd2 Qf4+, w g.

24 Nd2

 ( ): We arrive at a moment which

requires divatio ad magic as

much as ratioa aayss. ite's game teeters. How do we push him over?

 P.

24  Bxd5! 25 N xd5 N xe6! Leaves ear the back witch o f4 ruste ad swir; the air aroud her begs to cracke with power.

26 Qxa7

 ( ): The air is ripe with presentment of

immiet msfortue i the

regio of ite's kig. Back has access to tactics ad two ways to force the wi. Fid oe of them.

 Back rank.

26  Nc7! "J' adoube, aouces the embodeed kight, who operates uder the aegis of hs rook ad quee's authority. Back's hagg quee ist re ay hagg at a.

 #: Black has a seconary back rank trick with 26 ... N4! n White colapses.

27 Rxe8 Rxe8! 0-1 Due to 28 Nxf4 Re ate. White's kng lays own the buren of this fe as his spirit moves on to the next. Game 9 TVPetrosa-MBotvk Tranng atch, Moscow 1952 emi-lav Defence 1 4 N 2 c4 e6 3 Nf3 5 4 Nc3 c6 5 cx5

Petrosia prefers to carify the positio, rather tha chaege Botvk theoretica showdow with 5 Bg5 or 5 e3. 5  cxd5

il

a





Trasposg to the Exchage Sav, a virtuay uosabe positio for White  but as Petrosia discovers  this game, there are exceptios to every rue. Istead,  ... exd eads to the Exchage Variatio of the Quee's Gambit Deced, abeit a ess testg versio, sce White has aready committed a kight to f3.

: What fference oes that ake?

 The early f3-knight eployment cuts out set-ups ike e2-e3, B3, Nge2, f2-f3 an then later e3e4, which is generally consiere one of White' s more ngerous formations  QGD Exchange nes.

6 Bf4 Nc6 7 e3 NhS!?





Dreev's favourite e, perhaps Back's most combative ad aso most perious course. He attempts to chase dow the f4-bishop ad grab the bishop par, but at the heavy cost of weakeg ad riskg a ag  hs ow deveopmet. A more sedate deveopg set-up ike 7 ... Be7 8 Rc1 0-0 9 Bd3 Bd7 10 0-0 Rc8 11 Ne, as  Wag Hao-V.Maakhov, Cha vs. Russia match, Ngbo 2010, woud suit the traqui Petrosia just fe.

8 Bg5 Qb6 9 a3

9 Bb5 is White's mai optio;  e. g. 9 .. . Bd7 10 0-0 h6 11 Bh4 Bd6! (11 . . . g5?! 12 Bxc6 Bxc6 13 Ne5 Ng7 14 Bg3, V.Bhat-J.Becerra Rivero, US Team Champioship 2005, is dagerous for Back who caot grab o b2;  e. g. 14 ... Qxb2? 15 Nxc6 bxc6 16 Qa4! is awfu) 12 Nd2 Nf6 13 Nb3 Ne4!? 14 Nxe4 dxe4 15 Be2 Ne7! 16 Bg3 Bxg3 17 hxg3 Ba4 ad Back stood at east equa, V.Miov-A.Dreev, Iteret (bitz) 2004. 9  h6 After 9 . . . Qxb2?? the geie's boo proves iusory ad oses o the spot to 10 Na4, whe Back's quee breaks dow ito a hiccuppig sob.

10 B4 g5 11 Bg3 : Why not toss n 11 Ne5 - ?

 n some lnes b2 is tacticaly vunerable, especialy ones where White's queen gets iverte from . For example, 11 ... Nxe5 12 xe5 B7 13 Na4 (fter 13 Qxh5?? Qxb2, White has too any pieces hangng smultaneously) 13 ... Qa5+ 14 Nc3 gxh4 15 Qxh5 Rc, when Black enjoys the bishop par, evelopment lea an itiative.

11  Nxg3 12 hxg3 Bg7 13 Bd3 Qd8





The quee o oger serves a purpose o b6 ad returs home. We reach a cassic imbaace of sighty superior structure ad ope h-fie versus Back's bishop pair ad potetia o the dark squares. For ow, the positio remais cosed, so the bishop pair doest costitute much. But  the game ater opes up, they grow meagfu. 14 Nh2?! A head-scratchig momet. Petrosias pieces begi to behave irratioay, ike characters i a dream who say ad do the icomprehesibe.

: What is White's iea behn this bzarre retreat?

  suspect that Petrosian, ike yor fortnate writer, over-stuie Nmzowitsch n a misspent youth an hyper-fnesses n a position where he shoul be akng simple evelopng moves. Petrosian may have plne Qh5, followe by Ng4, when his iea works well. But with his contortion, he pays too high a price n ecentralzng a kight, senng it on to a mission to nowhere. What we see here is the positional player's isease, with which yor writer has also been afflicte his entre lfe: an obivious nere stiation of a position's ynaic potential. Botv, who suffere no such lness, s oon began a vigorous exploitation of Petrosian's eccentricity.

Morovic Feradez-A.Dreev, Moscow 2010, cotiued far more aturay: 14 Rcl Bd7 15 Nd2!? (hite stas, ervous about castg ad wakg to some sort of ... h6-h5-h4 attack) 15 . . . e5 ! (a good reactio to ite's ast move  Back opes the positio for his bishop pair) 16 dxe5 Nxe5, ad ow 17 Nf3 aga woud eave the positio dyamicay eve. 14  hS! Botvik aerty prevets Qh5. 15 Rcl Bd7 16 NS!? Yet aother eccetricity ad yet aother mssio to owhere. As is ite's custom  ths game, he is a bit too kee to weird it up. Petrosia, ued by the rigidity of the positio, correcty beieved ths fact gave him icese to maoeuvre to his heart's cotet. It was probaby high time to apoogize, ose face, cocede that his previous idea was weak, ad pay 16 Nf3. 16  Kf8!

Sidesteppg the vugar cheapo o d6.

: But at the cost of losng castng rights?

 After playng ... h6-h5, Black never ntene to castle kngsie, snce his rook now belongs on the h-file. Here f8 is a safe haven for Black's kng  n fact, far safer than castng.

17 Nfl?! The "to each hs ow phiosophy has its imits, ad here Pe trosia takes persoa sef-expressio a bit too far. The kight sighs, recocig himsef to hs ew, demoted post. To a cassica payer, Petrosia's hyper-refed fesses ad odd strategic gesticuatios may ook comprehesibe. ite's kights, obivious to dager, caper about ike chidre at pay at a picic. ite's diemma: cotue eve further dow a obviousy correct path, or submit to a humiiatg reversa of hs previous efforts? Petrosia opts for the former, dugg  a further cotortio to threate h5, but he forgets that positios fraught with tesio are ot coducive to such eisurey maoeuvres. 17  g4 18 Nd2 eS! Principle Open the position when yo own the bishop pair. Black hopes to knle s ome central heat.

19 Qb3 Back cotro the itiative after 19 dxe5 h4! too.

19   exd4 20 Nxd4 Nxd4 21 exd4 Qe7+

22 Kdl?! Petrosias ormay astoshg receptivity to the most mute strategic uace apparety took a exteded vacatio this game. The eccetricity-fest cotues ad White's wacky pieces wader about the board  haucatory fashio. Petrosia ca i afford yet aother extravagace whe he shoud busy himsef puttg his ramshacke house  order.

: Why i Petrosia's kng veer towa r the centre?

 He wante to activate his h-rook at e, at the cost of  nsecre kng  too high a price. His move is smply at os with the position's reqrements. He sho settle for 22 Kf Bx4 23 Qxb7 R8 24 Qx5 Bxb2 25 Rb Be6 with only a clear avantage to Black.

22  Bxd4 23 Rc7 23 Qxb7?? fais miseraby to 23 . . . Ba4+, pickg off the quee; whie 23 Qxd5?? is aso a o-go due to 23 . . . Bxb2 24 Rbl Ba4+ 25 Bc2 Rd8 with a crushg attack after 26 Qa5 b6! 27 Qxa4 Bc3. 23  Bb6 24 Re Qd6 25 Rxb7 Now Botvk fds a path to freedom for his kg's rook. 25  Rh6! The rook emerges, eager to participate.

26 Bb5 Be6 27 f4 Petrosia probaby wated to avoid the disma edg which arose after 27 Re2 h4 28 gxh4 Rxh4 29 Qb4 Qxb4 30 axb4. 27  gxf3!? A trade-off: Botvik opes the positio further but aows White's kight back to the game.

28 Nxf3 R8 29 Ne5 Q5

 ( ): Black threatens a huge check on c,

after which White's kig goes

for a ride. White ca pay 30 Qb4, agreeig to a ferior edig, or ese the startg shot 30 Rxf7+, which expoits a fork tactic o d7. ayse both ad decide carefuy.

30 Rxf7+?  f a person with a ark, violent past nvites us nto his home, we shoul be isnclne to enter. Petrosian, whose restrant vanishes when pitte aganst such temptation, isplays tactical ngenuity, but fortnately not enough to save him from himsef! Not all combnations shoul be playe. White shoul have entere the nferior enng with 30 Qb4! Qxb4 31 axb4 with some hope of savng the game.

30  Kg! A strage doube attack arses: Back threates the hagig rook o f7, as we as a deady quee check o cl. 31 Rf3? Petrosia had to appease the gods with sacrifice by tossg the exchage after 31 Nd3 Qd4 32 Rf3 Bg4. 31  Qc1 + Back seeks to force the extraditio of White's kg to face tria for past crimes.

32 Ke2 The kg receives his guest with cod differece. 32  Rc2+ Back's quee ad rook toss the kg's body aside, the way a racher woud dseased ivestock.

33 Kf1 Qd2 0-1 Back's major pieces eave a pume of misery aog White's secod rak. The white kg's diemma remds me of the Pk Foyd yrics: "Your ips move, but I ca't hear what you're sayg. I've become comfortaby umb.

: Why i White resign when he ha two ways to block on e2?

 Both fail iserably: 34 Re2 Rc1 +! or 34 Be2 Q4 !. Game 30 MBotvk-VSmyslov Worl Championship (2n atchgame), Moscow 1954 Nimzo-Indian Defence 1 4 N 2 c4 e6 3 Nc3 B 4 e3 b6 5 Ne2





White covers c3 agaist structura damage but cotorts a bit to do so. S 000 Ba6 Smysov immediatey puts his figer o the dowside to White's ast move ad pressures the ow hagig c4-paw.  ... c,  ... Bb7 ad  ... Ne4 are Back's ateratives.

6 a3 Be7 7 Nf4 More active tha the g3-square.

7  d5 8 cxd5 : This move compromises White's

castig rights. Ca he avoid osg them?

 f he wants, he can retan the central tension with 8 b3 0-0 9 Bb2, as n the earier game M. Bo tv-N. Novotenov, USSR Championship, Moscow 1951 .

8  Bxfl 9 Kxfl exd5 : t looks to me lke White ay have botche the openg.

He has a bad bishop o cl ad cat caste, meag his h-rook wi be out of pay for quite some time. Is ths correct?

 Not i you factor n White's next move. His rook ay actually be well place on h with his kngsie expansion plan. White ay also later play for e3-e4, as he oes n this game, which frees the c1bishop. The chances are probably closer to even.

10 g4!





If you work for a mob boss it's usefu to uderstad hs weakesses  but ever, ever metio them aoud. Now ths move is route, whereas at the time it was a startgy ew pa for hite, who attacks with a kgside paw storm. Hs oce out-of-pay rook ow ooks we posted o h. 10  c6 Smysov stabiizes his d-paw. Back's ma aterative is to ash out with 10 . . . g5!, which may  fact be Back's best bet  the positio; e. g. 1 1  Nxh5 12 gxh5 c6 13 Qf3 Qd7 14 e4 with tremedous compicatios, M.Nacu-J.Tait, correspodece 2002.

11 g5 Nfd7 11 ... Ne4 12 Nxe4 dxe4 13 h4, J.Redboe-Mechtg, Pardubice 2005, s aother possibiity. 12 h4 Bd6?! Back agrees to the dea  prcipe but hesitates to put it to writg ad sig by castg immediatey. The bshop uwisey diverts hie with a side ssue. The threat to damage hite's structure o f4 comes at the serious price of time, ever md that ite easiy avoids it with his ext move. Istead, Back shoud probaby eter the storm head o with 12 ... 0-0, ad 

White foows Botvk with 13 e4 dxe4 14 Nxe4, the 14 . . . Nc5! has a freeg effect o Back's game, T. Irzhaov-R.Rizzo, correspodece 2006. 13 e4! Botvik, at the cost of a isoai o d4, soves the secodary probem of hs formay active bishop ad augmets his piece activity further. 13   dxe4 14 Nxe4

: Does White risk overextension?

 He certay oes i he ishnles the position (which Botvk oesn't), but Black risks ner extension, which is a greater threat. Botvk wilngly takes on structral eficiencies n exchange for piece activity, which he now enjoys n abnance. At this stage Smyslov must have enre extreme frustration n his nabiity to locate a sngle fissre or crevice n Botv's seamless campaign of offence.

14   Bxf4?! The bishop egages  a exaggerated reactio to a trivia sight, ad fai to coect fair remueratio for his toi. Sometimes the medice itsef is the factor which ki the patiet. Smysov apprases the positio correcty, hopg to sove oe difficuty, but by dog so he takes o aother oe eve arger. Back, probaby ookg for pressure-easg swaps, agrees to a bad dea. Now White soo posts a kight o d6 with powerfu effect. Here 14 ... Bc7, retag some degree of cotro over  dark squares, was reativey best.

15 Bxf4 0-0 16 5 ReS Wakg to White's ext move, but 16 ... Na6 17 h6 g6 18 Qa4 improvemet.

17 Nd6 Eight beams of power radiate from the kight, ike spokes o a whee. 17  Re6 Back s uabe to survive 17 . . . Re7 18 g6! .

IS

o

 (  ): Botvk fon a way to see

the itiative. hat ca you come up

with i this positio?

 Principle Create conontation when your ponent lags in deelopment.

18 d5! The huma move.  #: Houdini ponts out that 18 g6! fxg6 19 hxg6 h6 20 5 is also quite awful for Black.

18  Rxd6! Back's oy chace. Smysov defty sidesteps immediate aiatio, the same way George W. Bush did whe the Iraqi jouraist shoe-thrower fug two smey projecties Bushward. Passive pay is fata here. After 18 ... Re7? 19 Qd4 c 20 Qc3, Back ca barey move ad h5-h6 is a horrific threat.

19 Bxd6 Qxg5 20 Qf3 Back's quee ad her odious sster o f3 siety regard oe aother with mutua detestatio. Botvk takes the steam out of Back's couterpay before it eve begs. A pure attacker ike Ta woud udoubtedy have retaed quees ad cotued with somethig ike 20 Rgl Qf5 21 Qd4 g6 22 hxg6 fxg6 23 Re. 20  Qxd5 Back's quee exhaes  costeratio ad agrees to the dea: a vasty ferior edg. To do otherwise s suicida. For exampe, 20 ... cxd5?? 21 Rgl Qd2 22 Rdl Qc2 23 Qxd5 Nc6 24 Rd2! Qa4 25 Be7! Nf8 26 Rxg7+ ! Kxg7 27 Qg5+ Ng6 28 Bf6+ Kf8 29 hxg6 s crushig.

21 Qxd5 cxd5 Back's kg ets out a happy sigh of reief. To him the absece of the white quee is fitey preferabe over her presece. Ufortuatey, the edg s ost for him as we.

22 Rc1

Threateg a back rak mate. 22   Na6 23 b4!





Back's offside kight s ess tha fattered by the owy roe assiged to  The parasite o b4 cotros its a6-host, who is dead, yet aive ad absoutey uder the paws sidious cotro. 23   h6 24 Rh3 Botvik activates his ast udeveoped piece. 24   Kh7 25 Rd3 Nf6 26 b5 Decidig the time has arrived to cash out. 26  Nc5 The castaway waves hs arms fraticay, oy to watch the cruse er recede over the horizo. 27 Bxc5 bxc5 28 Rxc5 Rb8 29 a4 Rb7





Basphemous thoughts cotue to bubbe up  the mds of Back's forsake pieces, which they dare ot utter aoud. The oe survivors of his desperate campaig are ecumbered with a porous defesive border.

 (  ): Come up with a concrete

pa to covert White's advatage to victory.

 Double rooks on the c-file to nvae the seventh rak, after which Black's efeners are nequal to the task of fenng off the queensie pawn majority, which soon rols forwar.

30 Rdc3! 1-0 Black re signs, beset with the nerstanng that he has little tme before White's ro oks overwhem hm on the queensie. Game   MBok-MTal Worl Championship (7th matchgame), Moscow 1961 Nimzo-Indian Defence 1 c4 Nf6 2 Nc3 e6 3 4 B 4 a3

: This move just oesn't appear logical, oes it? White expens a precious tempo, oly to force Black nto fictng aage upon the white structre.

 n a sense this ne is the epitome of ynamism. White wilngly hans over a tempo an alows amage, but he gets both bishop pair an strengthene centre n retrn. Toay, satisfactory efensive plans have been worke out for Black, but at the tme it was mysterious territory to most of Botvk's opponents. n this game, Botvk isplaye far eeper nerstanng of the openg, from which Tal never manage to recover equiibrium.

4  Bxc3+ 5 bxc3 b6 Today 5 . . . c ad 5 . . . 0-0 are more popuar.

6 £3

6  Ba6 Istead: a) 6 ... d5 eters simiar waters to Botvik-Capabaca (Game 8) from Chapter Oe, ad deed 7 cxd5 exd5 8 e3 0-0 9 Bd3 c 10 Ne2 Ba6 11 0-0 trasposes to the ote with 11 f3  that game. b) 6 ... Nc6 (gog drecty after ite's sore spot, the c4-paw) 7 e4 Ba6 8 e5 Ng8 9 Nh3 Na5 10 Qa4 Ne7 reachig a ubaaced positio with mutua chaces, A. Kotov-P. Keres, Budapest Cadidates 1950 .

7 e4 d5

The ast chace for 7 ... Nc6 which seems ike a superior way for Back to pay this positio.  cxd5 Bxfl 9 Kxfl exd5 10 Bg5 Retag cetra tesio fees stroger ituitivey tha 10 e5 Nfd7 11 Nh3 0-0, after which Back is ready for cetra couters with ... c7-c5, Meuders-R.Douve, Hiversum 1997. 10  h6 Botvik frowed upo ths move; but I dot much care for Back's game after 10 ... dxe4 11 Qe2 Nbd7 12 fxe4 h6 13 Bh4, F.Goda-R.Hirr, correspodece 1990.

11 Qa4+!? Botvik empoys a crafty psychoogica poy, correcty guessg that Ta's ove of compicatios ad dstrust of simpificatio woud be his ow worst eemy  the positio. Chess s't aways about payg the trsicay best move. Sometimes there is a best move for a give oppoet, which supersedes the overa best move.

: Why in't White wn a pawn with 11 Bxf6 Qxf6 12 ex5 - ?

 That is probably the ne Tal hope to enter. After 12 ... 0-0 Black's lea n evelopment somewhat compensates for the missng pawn. B otv writes: "True to my match tactics,  eclne the pawn sacrfice, so as not to concee the tiative to my opponent.

White ca aso specuate with 11 Bh4 dxe4 12 Qc2!? (eterg a strage Nimzo Backmar-Diemer hybrid) 12 . . . exf3 (perhaps decg is better: 12 . . . 0-0 13 fxe4 Ng4! 14 Bxd8 Ne3+ 15 Ke2 Nxc2 16 Rc1 Nxd4+ 17 cxd4 Rxd8 18 Nf3 c! 19 dxc5 Na6 20 cxb6 axb6 with at east equa chaces for Back, who has fewer weakesses to urse) 13 Re + Kf8 14 Nxf3 Nbd7 15 Kf2, whe White obtaed fu compesatio for his paw, G.Khodos-M.Mukhitdov, Novosibirsk 1962. 11  c6?!

Ta is uderstadaby ervous about chaegig Botvk i the edig. Nevertheess, he shoud remai resoute ad pay the best move: 11 ... Qd7! 12 Qxd7+ Nbxd7 13 Bxf6 Nxf6 14 Re! (stroger tha pushig forward, accordg to Botvik) 14 . . . Kd7 15 Nh3 with a microbe of a edge for White. I have a feeg Ta woudt have ost ths positio. 12 Bh4! Botvik, wsey distrustfu, ist fooed a bit by  oppoet's fasey beig maer. Sac'g a paw is much stroger tha grabbg oe with 12 Bxf6 Qxf6 13 exd5 0-0 14 dxc6 Nxc6, whe Back's ead  deveopmet more tha compesates. 12  dxe4 Somewhat forced, sce e4-e5 is  the air. Botvik gives this move a "?! mark ad writes: "A highy dubious decisio. The opeg of es  the cetre merey assists the deveopmet of White's itiative. But Kasparov very sesiby respods with: "It s easy to critice  but what shoud Back pay? I agree . The big questio: "Now what? fds o easy aswer. 13 Re g5 14 Bf2! Coverg d4 ad reforcg agast a future ... c6-c5 break. Botvk's move is more accurate tha 14 Bg3, which aows 14 ... Qd5!. 14  Qe7 Now 14 ... Qd5?! is met by 15 fxe4 Nxe4 16 c4 (oy possibe because of White's bshop o f2) 16 . . . Qe6 17 Qc2 f5 18 g4! ad Back fds himsef  dager of beg udermed o e4. 15 Ne2! The kight heads for g3, where it eyes both the e4-ad f5-squares. This s much stroger tha 15 fxe4 Nxe4! (this time Back gets away with the crime) 16 Qc2 f5 17 g4 Nd7 18 gxf5 Ndf6 ad Back ooks okay, despite the p o the e4-kight. 15  b5 16 Qc2

16  Qxa3!

: This looks crazy. Why woul Tal,

uder fire, take tme out for paw grabbg?

 t ay appear outlanishly greey but it's playe ner the theory: it's better to have somethng without nee, rather than nee somethng you on't have. Tal in't lke the looks of his position after 16 ... e3 17 Bxe3 Qxa3 18 h4 g4 19 Ng3, when Black is n eep trouble.

17 h4 gxh4?! Ta's pieces hod back ad m about, oe wig to be at the forefrot of the fight.

: Why i Black just allow the ecmation of his areay poor structre?

  on't nerstan the motivation behn the move, except to say: n a position where there are no goo answers, al moves are ba. Kasparov cale Tal's last move "negligent an recomene 17 ... g4, which woul be my choice, or 17 ... Rg8. n all cases, White has a firm grip on the avantage.

18 Bx4 Ta's opeg experimet goes horriby awry, much ke the scietst who buids a time mache, goes back 60 years, ad fa  ove with ad marries his ow gradmother, becomg his ow gradfather. Ta's aversio to edgs (especiay versus Botvik) costs him this game. Ta said afterwards: "For the frst time  my ife I was kocked out  the opeg! 18  Nbd7 19 Ng3 0-0-0 Back's kg rus away, tired of the fow of ite's reproaches dow the e-fe. There is o safe have for the kg across the board, but Ta cotues to cg to his shred of faith. 20 Nxe4 Rhe8!?

·



Ta attempts to mix it up with a piece sac. He s uikey to save himsef i the e 20 . . . Nxe4 21 Qxe4 Qxc3 22 Bxd8 Rxd8 23 Rxh6 Nb8 24 Rh4, whe ite's extra exchage shoud be decisive, despite Back's two coected passers. 21 Kf2! A great practica decisio, which deems materia cosequetia over itiative ad board cotro. Each of Ta's fratic attempts to compicate ru agroud o the reef of effectuaity, ad his hopes fa oe by oe. Istead, after 21 Nxf6 Rxe1 + 22 Kxe1 Qa1 + 23 Qd1 Qxc3+ 24 Kfl Nc 25 Bf2 a, Back woud s ti reta fshig chaces. 21  Nxe4+ 22 fxe4 f6 23 Ra1! ite's itiative grows to etha eve. 23  Qe7 After the rebuff, the quees dour siece makes everyoe aroud her ucomfortabe. Now she cocks her ear as she stes for the approach of eemy footsteps ear her kg's chambers.

24 Rxa7 Qxe4 25 Qxe4 Oe ca be a pragmatic reaist, ad at the same time beieve  mraces. Botvik's attack cotues, eve queeess. 25  Rxe4 Ta must have breathed a bit easier at this pot, havg forced quees off the board. However, Botvk had see deeper ad soo surges forward with attack duced a. 26 Ra8+ N8 Back is forced to the upeasat p, sce 26 ... Kc7? 27 Bg3+ ws o the spot.

27 Bg3 Kb7 28 Ral R8

·



What was oce thought to be a safe have is ow a bomb-scarred, taged abstractio of fragmeted meta ad cocrete. Back's kig is caught i a box, despite the comfortg absece of quees o the board.

 ( ): Fn hite's most accurate path to victory.

 The black kng's survival skils, accumulate over a long lfe of harship an eprivation, are stll not enough to save him.

29 R8a7+! 29 R1a7+ Kb6 (the kg, cowerg  hs coset, caps hads to ears, the souds of White's army's forced gress ubearabe to his ears) 30 Bd6 c 31 dxc+ Kc6 32 Rxb is sti w g but ot as s trog as the game cotuatio. 29  Kb6 30 Bxb8! The bishop adheres to b. 30  b4 Dazed ad dejected, Ta ca oy muster toke resistace. Botvik's pot: he procures a piece from hs abours sce the automatic recapture 30 ... Rxb?? rus to 31 R1a6 mate. 31 Bd6 bxc3 32 Bc5+ Kb5

 ( ): Black's kng loathes his

doctor's eforced caistheics. te to

pay ad force mate.

 The creatre, nconceivable n sze an scope, loos above the horizon, blottng out the entire sky. The first prnciple of the kng hnt: Cut off the enemy king, rather than give chase.

33 R1a4! 1-0 The other rook eivers ate on a n Tas kng floats bely up, a ea fish n a polute bay.

Chapter Four Botvinnik on Exploiting Imbalances I this chapter we study Botvik's acute receptivity to imbaaces o the chess board. The games which stick out to me  this chapter are the ast two: Meda Garcia-Botvik ad Matuovic-Botvik (Games 37 ad 38)  both variats of the Bc4 Pirc, ad both strikgy simiar  their expoitatio of opposite-cooured bshops whe attackg. Soo it becomes cear that Botvk's domatio of the dark squares  which his bishops  both games root out ike pigs siffg out truffes  heaviy outweighs White's cotro of the ight. As it s with most great strategists, ther imbaace aways shies whie the oppoet's grows irreevat. Game 32 M.Botvinnik-M.Euwe Word Champioship Touamet, The Hague/ Moscow 1 948 Semi-Slav Defen 1 d4 d5 2 c4 e6 3 Nf3 Nf6 4 Nc3 c6 5 e3 Nbd7 6 Bd3 Bb4?!





A book move at the time ad rather shady. : Why criticze this move? Black smply

pays the positio  Nimzo-Idia stye.

 The trouble is this is a Sei-Slav, not a Nmzo. Black place his pawns n a light square triangle on c6, 5 an e6. This means his ark-square bishop is of great value an shoul not be swappe off on c3. n a sense you are right: this is a Nmzo-nian, but a ba one for Black, snce he exchanges the useful move ... 0-0 (a position which John E covers n The Nimzo-Indian Me  Me) with the rather renant . . . c7-c6. : But Euwe in't swap it off on c3, i he?  True, but he i lose time, givng White the useful a2-a3 move for free.

7 a3 Ba Instead:

a) 7 ... Bxc3+?! 8 bxc3 scores a disma 17% for Back accordg to my database. b) 7 ... Bd6 aows 8 e4 dxe4 9 Nxe4 Nxe4 10 Bxe4, as  G.Kasparov-R.Hber, Brusses 1986. Now Back acks ... Bb4+, a move he woud ormay have access to  he hadt aready serted this earier o.

8 Qc2 Qe7 Back seeks a freeg ... d5xc4 ad ... e6-e5 cetra break. 8 ... 0-0 may just traspose. 9 B d2 dxc4 10 Bxc4 eS 11 0-0 0-0 12 Rael!

: Isn't this the wrong rook?

 It ooks ke the correct rook, based on the prncipe: Mass forces on your strong wing. White may at some pont trade on eS and then surge his kngside pawn ajority forward with f2-f4, e3-e4-eS.

12  Bc7 12 . . . e4?? rus to the tactic 13 Nxe4!  yet aother faw  Back's . . . Bb4 deveopmet scheme .

: If this is the case, why cn't Back

chop o c3 ad the pay . . . e5-e4 afterwards?

 Back's position st ooks rather sour to me after 12 . . . Bxc3 13 Bxc3 e4 14 NeS, foowed by f2f3, with a strong tiative for White. Note that 14 ... Nb6?? oses on the spot to 15 Bb4.

13 Ne4 Perhaps stroger tha 13 dxe5 which may be mstimed: 13 ... Nxe5 14 Nxe5 Qxe5 15 f4 Qh5, O.Karpeshov-M.her, Vogodosk 1983, whe White cat cotue thematicay with 16 e4?? sce it hags a piece to 16 ... Qc5+. 13   Nxe4 14 Qxe4 as! Deyg White ay future Bb4 cheapos. 14 ... Kh8?!, tedg ... f7-f5, ca be met with 15 Bb4! c 16 Bxc5! Nxc5 17 dxc5 f5 (17 . . . Qxc5?? oses to 18 Ng5) 18 Qd5. Now if Back tries to rega the paw immediatey with 18 ... Rd8?, White respods 19

Qf7 ad  19 ... Qxc5?? 20 Ng5! Back s hepess, sice White threates a smothered mate with 21 Qg8+! etc, as we as 21 Qh5 h6 22 Qg6! hxg5 23 Qh5 mate.

15 Ba2 Perhaps thikig about sidig back to bl at some future date. 15  Nf6 15 . . . Kh8! may be Back's best, sice it forces White to resove the cetra tesio.

16 Q4 The quee hopes to augmet her come with some freeace work o the kgside. 16  e4 Back must pick his dscomfort. 16 ... exd4 17 exd4 Qd6 18 Bbl ooks ike a bad isoai positio for Back, whose kg may soo be uder fire with so may white pieces gazg  the directio of the kgside . 17 Ne5!

 exceet paw sac.

17  Bxe5!? . . . which probaby shoud be deced.

: Why d o yo th Ewe accepted it?

 He may have feared White's kngside bid-p fter 17 . . . Be6 18 Bb1 Bd5 19 Bc3 Rae8 20 f3 exf3 21 gxf3, thogh this ooks better for Back thn what he got n the game. Another possibiity is 21 e4!? fxg2 22 xf6 Qxf6 23 Qxf6 gxf6 24 Nd7 with some advantage to White.

18 dxe5 Qxe5 19 Bc3 Qe7 19 ... Qh5?? woud be a regretfu decisio after 20 Bxf6. 20 f3! Dua prcipes foowed: Open the  game when leadin g in development, and also when  you own the bishop pair. For the paw, Botvik gets two bishops versus two kights  a ope positio ad the potetia for a kgside attack  more tha eough

compesatio. 20  Nd5?!

If a ay refuses to hep out, the at east make certa he doest hder. Overreactio s the great appeaser of og-carried fear. Back pays off White's importuate quee, just to get her out of his fe. After this move Back's etire word sps out of the pae of the sus eciptic cyce. I other words, a strategic misjudgmet o Euwe's part. He eimates White's attackg chaces for the heavy price of adg  a very difficut edg.

: What shod he pay?

 His best may be 20 ... Be6 21 fxe4! (probaby the move Ewe feared  this ne) 21 ... Bxa2 22 xf6 b5 23 Qg3! (Kasparov's sggestion) 23 ... b4 24 Be5 g6 25 axb4 axb4 26 xc6 f6 27 Bd4 Qxe4 28 Rb6 Bd5 29 xb4 with better chances to hod than i the game.

21 Qxe7 Nxe7 22 fxe4 b6?! Back frees his a-rook from babysittg duties o the a-paw at the cost of weakeg his queeside. Botvik, Euwe ad Kasparov a abeed this move a error. Ateratives: a) 22 ... Ng6 (Euwe's suggestio) 23 Rd1 Be6 24 Bxe6 fxe6 25 Rd7 Rxfl + 26 Kxfl Rf+ 27 Ke2 Rf7 2 Rd6 a4 29 Rd4 b5 30 Rd+ Rf 31 Rd6 doest ook saveabe for Back. b) 22 ... Be6! (Botvik's suggestio, ad the best try  my opio) 23 Bxe6 fxe6 24 Rd1 Rfd (Back must cover agast the threateed vasio of the seveth rak) 25 Rxd+ Rxd 26 Bxa5 with a extra paw. Ths e may be Back's best hope sce White's kgside majority s somewhat hobbed by the doubed e-paws.

23 Rdl Ng6 Back remas uder heavy pressure after 23 . . . Bg4 24 Rd6 Bh5 25 Rd7 Rfe. 24 Rd 6 Ba6 25 Rf2 Bb5 26 e5! The e-paws, evi fueces, soo maage to destroy a virtue  Back's positio. Botvik cotuay earges upo his previous strategic gas by addg the e5-e6 rammg threat to his edger;  e. g. 26 ... Rae 27 e6 fxe6 2 Bxe6+ Kh 29 Rxf+ Nxf 30 Ba2 ad Back ca barey move.

26  Ne7 There is o eed for a homeess waderer to make haste. What is the rush whe your destatio is owhere ad aywhere?

27 e4 Botvik seeks to cut out a ... Nd5 ideas. However, Back's positio is competey hopeess if White simpy seizes ad impouds the seveth rak with 27 Rd7! Nd5 28 e6! Nxc3 29 exf7+ Kh8 30 bxc3. 27  cS War's upheava created disocatios amog Back's citizes, ad severa pieces ad paws fee out of syc. Back's fratic motios covey a sese of dre urgecy. 27 ... Rae8 is o improvemet due to 28 e6 f6 29 Rd7 ad Back  paraysed.

28 e6 The ead e-paw creates a deep wedge  Back's positio. 28  £6 Not 28 ... fxe6? 29 Rxe6 Rxf2 30 Kxf2 Kf8 31 Rxb6 ad Back ca resig.

29 Rxb6 I the heat of batte, the oss of comrades fees ike a amost commopace evet. 29  Bc6? Truth, which appears before us,  aways the destroyer of deusio. The ar fees saturated with the essece of immet vioece. Back just budered  a hopeess positio. Now the archer awaits, thikg: "Come coser, just a itte coser ad I have you withi arrow-shot rage. You oy get oe shot. The positio  craked to maximum tesio. Now  the time to take aim ad fre.

 ( ): How did Botvk forcefy end the game?

 Defection. The ramshacke constrct coapses, as we a knew it wod.

30 Rxc6! The rook emerges, crashig dow sideways, ike a bue mar breakig water. 30  Nxc6 The icapacitated kight  ot up to the task of hatig White's itet. 31 e7+ Rf7 The wayward rook i despair, refectg upo Back's may ss, asks himsef: "What have I become? 32 Bd5! 1-0 This was a breakthrogh frst wn over Ewe, previosy Botvk's boogie an. If yo are gong to break a osng streak, then the optimm tme to do so is n a word championship tournament! Game 33 WUhl-MBotvk Much Oympiad 1958 Nimzo-Indian Defence 1 4 e6 2 c4 Nf6 3 Nc3 B 4 e3 b6 5 B3

Botvik, perhaps the greatest Nimzo-Idiaer of a time, was fuet with Back's side as we. I my opio White's best chace at a edge ies  5 Ne2, which Botvk payed agast Smysov  Chapter Three (see Game 30) . 5  Bb7

6 Nf3 I dot trust 6 Ne2 Bxg2! 7 Rg1 Be4 for ite;  e. g. 8 Bxe4 (the rook gets corered after 8 Rxg7? Bg6) 8 ... Nxe4, ad ow if ite ssts o regag hs materia with 9 Rxg7?, he waks to 9 ... Nxf2!, H.Restifa-D.Ada, Bueos Aires 1992.

: Why doesn't White go for 6 f3 to bid p for a big centre with e3-e4 - ?

 Back won't give hm the tme. For exampe, after 6 . . . cS 7 Ne2 cxd4 8 exd4 0-0 9 0-0 dS, White gets a passive version of the isoani snce his knight sits on e2 rather than on f3. He can't opt for hangng pawns either, snce 10 b3?! is met by 10 ... dxc4 nd White is forced to recaptre with his bishop, R.Ligori .Fedorchk, Porto an Giorgio 2008.

6 000 Ne4 7 0-0!? hma offers a paw. 7 000 fS! Botvik  uike Lasker, Korcoi ad Fscher  was ever much of a paw grabber. He hed a deep aversio to cedg itiative, eve temporariy, for materia of ay kd.

: Does White get enogh compensation

if Back goes ahead ad accepts the sac?

 Accordng to my database White scores above average after both 7 ... Bxc3 and 7 ... Nxc3.

8 Qc2 Bxc3 Fay, Botvk gives up a bishop for a kight to fict damage upo ite's structure.

9 bxc3 0-0

lO Rbl A rare e.

: Why does White post his rook on a fe withot targets?

 He ntends to undobe his pawns with c4-c5  which B otv d oesn't aow. o, essentiay,

a) 10 Ne1 c 11 f3 Nd6 12 Ba3 Na6 13 Qe2 Qe7 14 e4 fxe4 15 fxe4 Rxfl + 16 Kxfl e5! ad I prefer Back's chaces i this rigid structure, A.Graf-G.Kasparov, Geeva (rapid) 1996. b) 10 Nd2 Qh4 11 f3 Nxd2 12 Bxd2 Nc6 ad Back achieved harmoious deveopmet, V.mysov-D. Brostei, Moscow 1956. 10 ... cS Bockig White's itetio.

the move accompishes nothng. Instead:

11 a4 Pag a4-a5 ad axb6, creatg a target o b6. 11 ... Qc7 12 as d6 13 Nd2 Nxd2 14 Bxd2 Nd7 1S Rb2? White shoud exchage o b6. The text has a way of aduteratg a aready effective pa ad the hoped-for augmetatio of his queeside buid-up fai to pass. With a sge careess move hma doubes his difficuties ad quadrupes the bra power requiremets to extricate himsef. Botvk ruthessy expoits what see at first just a ty accuracy.

15  bxa5! Back's strategic threat s ... Nb6 foowed by ... a5-a4, so ite must hurry to rega his paw. 16 Ral?! Uhma was afraid to eter a rather sour-ookg edg after 16 e4 Bxe4 17 Bxe4 fxe4 18 Qxe4 (meacg e6, as we as Rb7) 18 . . . Nb6! 19 dxc5 dxc5 20 Qxe6+ Qf7 21 Qxf7+ Rxf7 22 Ra2 a4, but ths may have bee ite's best bet to save the game. 16  Nb6! 17 Rxa5





The discovered attack 17 . . . Nxc4?? fais for Back, sce it hags two pieces for a rook. Nevertheess, Back's positio throbs with potetia ad sweet dreams of puder.

 ( ): How did Botvik get around the

probem ad maage to fict

the discovery  aother way?

 The meddng bishop decides to ntervene n the atter. Botvn emnates White's best defender to underne c4.

17   Be4!! The pseudo-sac has the effect of fictg upo ite a state of cofused dsequiibrium ad arrives as a jarrg ed to Uhma's previous fow of

traquiity.

18 Bxe4 The od priest o d3 crosses himsef ad mutters a protective prayer, wary of the uatura creature passig by o e4. Not 18 Ra6 Nxc4! ad hite's bshop fubs the missio of defedig c4 big time. 18   fxe4 19 Qb3 Botvik's cever poit is that, after 19 Qxe4? Nxc4 20 Qxe6+ Qf7!, Back's quee deas with her sster with the casua ease of a woma swattig a mosquito that dares to make a uisace of itsef. Back ws the exchage with a easy coversio. 19  Nxc4!

Discovered attack. Back's kight emerges with terribe purpose. Botvik is dsmissive of Uhma's attempts to hag o to c4, so as to seem amost offhad ad derisive.

20 Qxc4 Fay removg that hated kight, who sat so smugy o c4, but for the high price of the exchage. So hite's quee acquiesces to the cocessio, vowg: "This is the ast time! Ufortuatey, her predictio turs out to be correct. 20  Qxa5 The a-rook is remaded to discipary actio for earier discretios. 21 Qxe6+ Kh8 22 Ra2 Qc7 23 Qxe4?? The white quees eyes soo we up with remorse at her mstake. Uhma, by ow competey demoraized, budered  a hopeess positio.

 ( ): How did Botvk end the game n a snge move?

 Dobe attack.

23 000 Qf7! 0-1 After 24 Be White's kng weeps with gratitde that his fe is spared. Bt the ceebration turns ot to be preature snce 24 ... Qxa2 picks off a rook. Game 34 VSmyslov-MBotvk Word Championship (1st atchgame), Moscow 1958 Caro-Kann Defence

Botvik had a deep, feong respect for his opponent. In 1983, mysov, n his sxties and we past his prime, iracuosy qafied for the Word Championship Candidates matches, noray the preserve of the young, psettng GM Zotan Ribi n his first match. He drew his next (and then advanced on a cky con toss) aganst the heavy favoured German GM, Robert Hner. Then he faced by far his most fearsome threat, Garry Kasparov. I remember top payers predictng that the ageng mysov woud be unabe to score a snge haf pont aganst the young sperstar. omeone asked Botv f he thoght mysov woud get crshed n the atch. Botvk's stern response: "One does not crsh mysov. In the event, mysov gave a good account of himsef bt ost 8/-4/ aganst the unstoppabe force of Kasparov.

1 e4 c6 2 Nc3 d5 3 Nf3 The famous Two ights variatio of the Caro-Ka reached a eve of otoriety, may due to Fischer's repeated, disastrous osses with it  the 1959 Cadidates Touramet. 3 000 Bg4 4 h3 Bxf3 The frst imbaace appears. Back gives up bshop for kight.

: Why hand over the bishop par withot a fight?

 Back pans to set p a ight square pawn tringe on c6, d5 and e6, so he ogicay hnds over his technicay bad ight-sqared bishop, to rean with a snge good dark-sqared bishop. The oss of the bishop pair at this pont doesn't constitte a rea probem for Back snce the position for now reans cosed, and his kights deem themseves on eqa footng with White's bishop par.

4 . . . Bh5!? is the psychotic cous. Pay may ru 5 exd5 cxd5 6 Bb5+ Nc6 7 g4 Bg6 8 Ne5 Rc8 9 d4 e6 10 Qe2 Bb4 11 h4 Nge7 12 h5 Be4 13 f3 0-0  a theory so far, to

reach this irratioa positio. I suspect White is sighty favoured  the comg compicatios.

5 Qxf3 Nf6 6 d3 e6

7 Be2 Aternativey: a) 7 Bd2 (the move favoured by most FIDE 2600+ payers, though it gives Back few difficuties) 7 ... d4 8 Nd1 c 9 g3 Nc6 10 Bg2 h5 11 h4 g6 12 Qe2 Bg7 13 f4 Qb6 14 b3 Ng4, whe ... Ne3 is  the ar ad Back achieved dyamic equaity, F.Vaejo Pos-A.Motyev, Germa League 2012. b) 7 g3 (Bobby Fscher's ufortuate favourite here) 7 . . . Bb4 8 Bd2 d4 9 Nb1 9 . . . Bxd2+ (or 9 . . . Qb6 1 0 b3 a 1 1 a3 Be7 12 Bg2 a4 13 b4 Nbd7 14 0-0 c with exceet queeside couterpay for Back, J.Fscher-P.Keres, BedZagrebBegrade 1959) 10 Nxd2 e5 11 Bg2 c 12 0-0 Nc6 13 Qe2 g5! 14 Nf3?! (14 f4 was ecessary to free White's positio) 14 ... h6 15 h4 Rg8 16 a3 Qe7 ad Back may aready be sighty better  a reversed Kig's Idia-stye positio, R.J.Fischer-T.V.Petrosia, Bed ZagrebBegrade 1959. 7 000 Nbd7 Hatg e4-e5.

: Isn't Back spposed to pay 7 ... Bb4 at this pont?

 Both moves are possibe. One of my stdents experienced dfficties from 7 ... Bb4 8 e5!? Nfd7 (8 . . . d4! ? 9 a3 sighty favors White) 9 Qg3, which is anoyng for Back bt ay be okay after 9 . . . d4! ? 10 Qxg7 f8 11 Qg4 Nxe5 12 Qe4 Qd6 13 a3 Ba5 14 b4 Bc7.

8 Qg3 White's idea s to exert pressure o g7.

: How doe s B ack compete his deveopment?

 ee his next move!

8  g6!

: Bt d oesn't this create te rribe punctures aong the dark sqares?

 There is no basis for nervosness n this position. Back keeps contro over his dark sqares here, any snce he reans with a dark-squared bishop.

9 0-0 : How abot 9 Bf4, ntendng to meet . . . B g7 with Bd6 - ?

 Your pan ooks dangeros, bt Back need not pay his bishop to g7. He can counterattack with 9 . . . Qb6 10 0-0-0 Bb4 11 e5 Bxc3 12 bxc3 Ng, when he is ndeed weakened aong the dark sqares, bt st generates sufficient counterpay aganst White's kng and weakened qeenside pawns. For now Back's knights hod their own aganst the bishop pair, snce the position remans cogged.

9  Bg7 10 Bf4 10 Qd6 Bf8 11 Qg3 Bg7 woud't bother Botvik a bit, sce every draw game with Back  a Word Champioship match represets a mor victory. 10   Qb6 11 Rabl 0-0 12 Bc7?! As it ater turs out, this move does more harm to himsef tha good. 12  Qd4!? I have the feeg Botvk eggs hs oppoet o, tautg  by cetrag hs quee. Curiousy, White has o obvious method of expoitg the quees positio. 13 Bf3?! The bishop shoud retur from its misadveture with 13 Bf4. 13  e5!

Seg hs far share of the cetre.

: Bt doesn't the move aso oosen his centra pawns a tad?

 The nte oosenng of Back's centre is overshadowed by the fact that he aso cts the d6bishop off from the kngside, forcng it to a3, where it ooks a bit rreevant and somewhat vnerabe.

14 Bd6 Smysov ist gog to fa for 14 Ne2?! Qc5 15 c3?? dxe4 16 dxe4 Rfc8 17 b4 Qe7 18 Ba5 b6, whe hite's bshop gets igomiousy trapped. 14  Rfe8 15 Ba3 dxe4 Botvik decides the timg s right for reease of cetra paw tesio.

16 dxe4 : hodn't White recaptre with a

piece to ope the positio for hs bishop par?

 By dong so, he wod hand over a greater share of the centre; i.e. 16 Nxe4 Nxe4 17 Bxe4, when Back ooks fne after 1 7 . . . Nf, ntendng .. . Ne6.

16  b5 Grabbg space ad payg upo the bshop's strage positio o a3 by threateg the cheapo . . . b5-b4. 17 Rfdl Qb6 18 b3 Nc5!

·



Dua purpose . Botvik graduay improves the positiog of his pieces ad: 1. He reews the . . . b-b4 threat by coverg a4. 2. He pas to sk the kight to d4, via e6. Of course 18 . . . b4 fai to w a piece after 19 Na4.

19 Bl I aready prefer Back's game, due to his growg gp over the cetra dark squares.

: Even thogh White owns the bishop pair?

 Correct. White's bishops fa to impress n this cogged po sition.

: What is so wrong with White's position?

 Not mch right now, bt yo fee the degeneration abot to happen with itte thngs. For nstance, White's ot-of-pay qeen canot be easiy rosed from her now White-ish coma where she ies. That, together with the erodng natre of White's dark sqare contro, eads me to favor Back. White's correspondng contro over the ight sqares jst isn't there to compensate.

19  Qc7 Sidesteppg the comg Be3 p.

20 Be3 Ne6 21 a4 a6 Botvik prefers retag the tesio over the e 21 . . . b4 22 Na2 c 23 c3. 22 b4!? is paw turs out to be a future weakess. Somehow White's positio cotues to de geerate ad it's difficut to see just why. 22  Rad8 23 Be2 Qe7 24 axb5 axb5 25 Rxd8 Rxd8

26 Bb6 Of course the ave 26 Qxe?? hags a piece t o 26 . . . Nh. 26  Ra8 27 f3?! Smys1v, axious to reitroduce hs quee to the game, further weakes the kgside dark squares. Houdini suggest the rather awkward 27 Ba as White's best move. 27  Ra3 28 Qe1 28 Ndl?? drops materia to 28 ... Nxe4!. 28  Bh6 29 Bf1?! Wastg time. The bad bishop, of choeric temperamet ad dispositio, does't take kdy to the abe "bad whe referred to himse. 29 Ndl was more accurate. 29   Nd4 30 Bc5 Qe6 31 Bd3 White's gagy bad d3-bishop appears to be the big, dumb kid who fai two grades ad ow ooks ridicuous packg  fifth grade behd hs thrd grade desk. Meawhie, we sese more dissoace from White's side, the way a high schoo bad competety attempts to piece together a difficut Bartk orchestra piece.

 (  ): Fnd a pan which dramaticay

icreases Back's growig advatage.

 A convenient expedient presents itsef to generate a new and favorabe mbaance: force opposite-coored bishops.

31 000 Nd7! Esurig the swap. 32 Bxd4 exd4 33 N e2 Be3+ The tresome, kow-it-a bishop eucidates the obvious: he rues the dark squares.

34 Kl Ne5 The kight estabshes residecy o e5, simiar to the oset of a migraie.

: Why does the presence of opposite-coored bishops hep Back?

 Back donates the dark sqares, whereas White's mopng pieces do anythng bt donate the ight sqares.

35 Qfl Qd6 Now d3 reay is uder attack. The growig dark square pague brigs o fever, bois, deirium ad evetuay death to the ifected. Botvik avoids the shaow trap 35 ... Nxd3 36 cxd3 Rxd3? 37 Nf4.

36 f4 Compete desperatio. 36 Nc1 Rc3 37 f4 Nxd3 38 Nxd3 (38 cxd3 Qxf4 s o soace) 38 . . . Qe6 39 Qf3 Qa2 40 Rd1 Qxc2 41 Qf1 Bd2 is eve more hopeess. 36 000 Nxd3 37 cxd3 Rxd3 Botvik sags a key paw, creatig a passed d-paw, ad a the whie retag a hs activity advatages. 38 Qf3 Rd2 39 Rf1!

Prayig he ca stir somethig up o the f-fie. Up to here ite's pieces foated aimessy, chuks of vegetabes simmerig  a broth. But ow he prepares a ast ditch effort to reach Botvik's kg. 39  Qxb4 40 e5 Hopg to have time for Ng3 ad Ne4, whe he may be abe to geerate kgside threats . Hope of a happy edg, a bright future  o matter how uikey  has a way of trasformg a ubearabe preset to a sighty more bearabe oe. 40 f5 Qc4 covers everythig. 40   Qc4 41 Ng3 If a six-year-od is give a choice betwee a chocoate chip cookie ad a sice of sashimi, I'm guessg she wi go for the cookie. I the same way, Smysov refuses to submit to passive defece ad pas a fa uge  a positio where his pieces fumbe about  dsarrayed squaor. If give time for Ne4, f4-f5 ad e5-e6, some thg may happe.

 (  ): How did Botv greaty dish his opponent's

ofty attackg ambitio

ad que the surgecy eve before it bega?

 Force a swap of ro oks via c.

41  Rc2! 42 f5 Rcl The rook cears his throat with " Ahem, attemptg to get the master's attetio o fl. 43 e6!? The hoped-for crushig shot just st there. 43  fxe6 44 fxg6 44 f6 is effectivey quashed by 44 . . . 7. 44   Rxfl+ 45 Nxfl hxg6 46 Qf6 46 Nxe3 dxe3 47 Qxe3 b4 is hopeess. 46  b4 47 Kh2 g5

·



Forcig more simpificatio. Now Smysov's oy prayer is a uikey perpetua check. 48 Nxe3 dxe3 49 Qxg5+ ite's ueviabe state of destitutio eaves him with itte choice but steag to feed hs hugry famiy. 49  Kf7 50 Qxe3 b3 51 Qe5 cS Here they come. ite's dream of perpetua check fais to materiaize. 52 Qc7+ Kg6 53 Qb8 Or 53 Qg3+ 5 ad  54 Qf3+ ? the 54 . . . Qf4+ (check! ) forces the quees off the board. 53  Kf5! Back's kig has too may hidig paces ad a perpetua check just does't exst. 54 Qf8+ Ke4 55 Qf6 Qd5! The surrogate mother, as stipuated i her cotract, must had over the fat upo birth, yet refuses to do so. Botvk's quee jeaousy guards her property, whie makg room for her kg to dace his way out of the checks. 56 Qf3+ Kd4 57 Qdl + 57 Qf4+ Qe4 58 Qf2+ Kc4 59 Qf1 + Kb4 dodges aga. Each time the back kg is haded a check-free move, a paw advaces earer promotio. 57   Ke5 58 Qe2+ Kd6 59 Qa6+ Ke7 60 Qa7+ Kf6 The kg edges away from the amorous quees advaces with a strog degree of revusio. 61 Qh7 Qe5+ 62 Khl b2 0-1

·



Back's qeen midwfes the pawn to birth on b after Back's kng escapes a barrage of checks. mysov reaizes the ftiity of payng ot a ne ike 63 Qh8+ Kf5 64 Qh7+ Kf4 65 Qh4+ Ke3 66 Qe + Kd4 67 Qd2+ Kc4 68 Qc2+ K4 69 Qd2+ Qc3 70 Qf4+ (White, n his endess attempts to sta and fiibster, ony anages to rean fiibsted) 70 ... Qd4 7 Qb8+ Kc3 72 Qg3+ Kc2 73 Qg6+ Qd3 ends the checks, as we as the game. Game 35 MBok-MTal Word Championship (3rd matchgame), Moscow 1961 Nimzo-Indian Defence 1 c4 Nf6 2 Nc3 e6 3 4 Bb4 4 e3 0-0 5 B3 5 6 a3 xc4

This is a side ie, as is 6 . . . Be7. Takg the kight coud traspose to Botvk Capabaca  Chapter Oe; i.e. after 6 ... Bxc3+ 7 bxc3 c 8 cxd exd (see Game 8), though Back might ao pay 8 . . . Qxd!? or, more usuay, 7 . . . dxc4 8 Bxc4 c.

7 Bxc4 Bd6

: Isn't Back wastng tme with his bishop?

 Back does agree to a tempo oss n this ne, bt remember, White ost time movng his c4bishop twice as we. Back pays for an ... e6-e5 break, sighty offbeat for a Nmzo-Indian bt qite sound.

8 Nf3 Nc6 9 b4!? 9 0-0 e5 10 h3 (10 d5 Ne7 s payabe too) 10 ... Bf5 11 d5 Nb8 gives Back a reasoabe positio sice he cotros e4, G.Kuzm-B.Guko, USSR Champioship, Tbiisi 1978. 9  e5 10 Bb2 Bg4 11 d5 White seizes a itte cetra space. Two games ater, Botvik deviated with 11 dxe5 Nxe5 12 Be2 Qe7 13 Nb5 Rfd8 14 Qc2 a6 15 Nxd6 cxd6 with a edge to White due to hs superior structure ad bishop par, M.Botvk-M.Ta, Word Champioship (5th matchgame), Moscow 1961. 11  Ne7 12 h3 Bd7 12 . . . Bf5! is a possibe improvemet, Aav-Khoup, correspodece 1994. 13 Ng5! Botvik fights for cotro over e4. 13  Ng6?! Ad here 13 ... a! was correct.

 ( ): The normay tacticay aert Ta isses a hidden

trick  the positio.

How did Botvk sigificaty improve his positio?

14 Ne6!  Dobe attack.

: Where is the dobe attack?

14  fxe6 15 dxe6  ight here. Back mst return the piece, sbittng to an ferior position, snce White threatens both the d7-bishop and, far more seriosy, e6-e7+, wng Back's qeen.

15  Kh8 16 exd7 Back's separabe bishops, uexpectedy become separated. 16   Qxd7 17 0-0

who

previousy

traveed



cofederacy,

·



ite has ceary got the better of the dea: 1. He picked up the bishop par. 2. ite has the superior paw structure. 3. ite cotro the ight squares ad potetiay a hoe o e4. 4. Back has a bad remag bishop. Noetheess, Ta remas dagerous sce may of his pieces urk meacgy aroud hite's kg. 17 000 Qf5 18 Nd5 I prefer 18 Qbl! (sti fightg for the key e4-square) 18 ... e4 19 f4!, which pretty much eds Back's attackg stabs o the kgside. 18 000 Ng8?! Back's kight, teder of the roya stabes, dreams of a ife greater tha just merey that of a shoveer of horse maure. Ta dispays a marked procivity toward extremes ad,  dog so, waks precariousy cose to the precipice of passivity. Payed uder the phiosophy:  we cat fd truth, the et's vet our ow  much the same way a dsreputabe research scietist fudges data to match a dubious hypothess. After this ucharacteristicay passive/aggressive respose, Botvik stregthes hs positio sigificaty.

: Isn't it nsne to retreat n this sitation?

hat is the motivatio behd the move?

 If a sane person meets one who is nsane, does there not at east exist an eqa poarity of opnion? From Tas perspective, his move is competey natura! Han fraity dispays itsef most when we try and tame natura forces to our kng. Ta ideaisticay soght to retan pieces on the board, bt retreatng n an open position is not a good way to do it. As he ater discovers, the trobe with his ideaism is that it fas to hod p under the stern test of Botv's practicaity.

19 Qg4 Qc2 Ta has itte terest  defedg a ferior edg after 19 ... Qxg4 20 hxg4 e4 21 Radl.

20 Qe2 Qf5 21 Qg4 Botvik, a ifeog time pressure addict, ever mssed a opportuity to toss 

a repetitio to get himse a coupe of moves coser to the move 40 time cotro. 21  Qc2 22 Qe2 Qf5 23 e4!

Nyet to the draw! 23  Qd7 24 Rad1 Rad8 24 . . . Nf4?? waks to 25 Nxf4 Rxf4 26 Bxe5.

25 Qg4 Hi! I'm back aga. 25   Qe8 26 g3 Botvik cotuay improves his positio, cuttg off access to both f4 ad h4. Now Back's kights are cosiged to a ife of dreary eui ad overpowerg mediocrity. 26  Nh6?! A waste of time. 26 ... a woud at east represet a sma dstractio to ite o the other wg. 27 Qh5 Ng8 28 Qe2 N6e7 29 Ne3! ite's kight is better paced tha either of Back's, so ow it is he who avoids a swap. 29   Nh6 30 Ng4! Now he does aows a kight exchage sce it stregthes his paw structure. 30  Nxg4 31 hxg4

Ta has bee thoroughy outpayed strategicay: 1. White has the bishop pair i a semi-ope positio. 2. Back edures a bad bshop. 3. White has a grip o the ight squares. 4. White has access to a ope h-fie, ad may cotiue with Kg2 ad 1, auchig a future kigside attack. 5. Back must watch out for f2-f4 ideas ater o. 31  Nc6 Ta's defece is based o ... Nd4, which woud resut  opposite-cooured bshops. But as he ater discovers, ths  o way heps his positio, sce opposite cooured bishops favour the attacker  White.

32 Kg2 Now Rh1 s  the air. 32   Be7 33 B d5 N d4!? Compete desperatio. Ta must make a excruciatg decisio: expesive submissio or a expesive uge at depedece? I either case it matters itte sce either path saves  34 Bxd4 exd4 35 Bc4! A typica Botvik decisio, choosg itiative over materia. Botvk's ust for power wi ot be pacated with tribute of a mere paw. He bockades, preparg a fa assaut o Back's kg, rather tha distractedy chase after the b7-paw. 35  cS 36 b5

Principle: Opposite-coloured bishops favour the attacr. This mbaance proves to be Ta's undong and his

1. His two coected, cetra passers are useessy bockaded, ad therefore egated. 2. White's kgside majority soo ros forward toward Back's kg. 3. I cojuctio with the ope h-fie ad opposite-cooured bshops, this spes doom for Back's kg. 36  B£6 37 £4 d3 Some chieftas attempt to cotro a uruy popuace ot through raw miitary might, but through the subte dog out of privieges. The bshop's shift to d4 does itte to aeviate his kg's troubes. 38 Rxd3 Rxd3 39 Bxd3 Bd4 40 e5 Targetg h7. White's itiative fows  a ceaseess stream of threats ad uremittg strategic pressure. It amost fees ike White s two paws up, sce hs 4:2 kgside majority advaces uopposed. position remans weighed down with mutipe, dreadfu strategic woes:

40   g6 41 Rh1 The cairvoyat rook gazes ito the opposg kg's future ad  puzzed to fd oy a back void. 41   Kg7 42 Qe4 b6 43 Bc4 1-0

: I reaize Back is n bad shape, bt

ist this a premature resigatio o Ta's

part?

 When the back death ravaged Europe n the 14th century, it wiped ot 100-200 mion peope. Famy members of a victim wod franticay try and keep their stricken oved one awake and aive. In most cases, the page proved stronger than the faiies' deternation. It is the same way here. A feeng of hepess ftity for his kng's fe coours Ta's every perception. He understands it is jst a matter of time before his dazed monarch, stricken with some form of cataepsy, wi expire. Ta's position progressed we beyond savageabe eves. For exampe: 43 ... Qe7 (Back's qeen asks her partner "what w become of s?  bt the kng refses to answer and merey tgs nervosy on his beard) 44 g5 Bb2 45 f5! ! (the cearance sac gives White's qeen access to h4) 45 . . . xf5 (45 . . . gxf5 46 xh7+! eads to pretty mch the same thng) 46 xh7+! (i yo are hungry and have the money, why order a frga mea?) 46 ... Kxh7 (now the kng is bffeted abot unti his consciosness fades to back) 47 Qh4+ Kg7 48 Qh6 ate. Game 3 MTal-MBotvk Word Championship (18th atchgame), Moscow 1961 Caro-Kann Defence  e4 c6 2 4 5 3 e5 Bf5 4 h4

·



I my book o the Caro-Ka, I abe this ie a "dagerous optio .

: This amost appears a random move. What is the pont?

 This pont is that f Back contnes noray with 4 . . . e6??, he waks nto the trap 5 g4! Be4 6 f3 Bg6 7 h5, wng the bishop. o far, 56 ufortunates have entered the trap accordng to my database. (Over the years at east haf a dozen of my stdents have faen for it as we.) Even more amazng is the fact that three times, White didn't know how to wn the piece om this position!

4  h6 I The Caro-Kann Move by Move I advocate 4 ... h5 as possiby Back's best ie;  e. g. 5 c4 e6 6 Nc3 Ne7 7 Nf3 Bg4 8 Bg5 Qb6 9 Qd2 reaches a messy positio with chaces for both sides, E.Aekseev-P.Ejaov, Europea Cup, Eiat 2012.

5 g4 Bd7 : Why not to h7, the natura sqare?

 A strategic trap. White gets terrfic compensation for the pawn after 5 ... Bh7 6 e6! fxe6 7 Bd3 Bxd3 8 Qxd3, when the qeen nvades on g6.

6 c3 6 h5 is probaby White's best shot at a edge; e. g. 6 ... e6 7 f4 c5 8 c3 Nc6 9 Nf3 Qb6 10 Rh2!? (a cever method of coverig b2 to eabe the dark-squared bishop to deveop; Ta himsef tried 10 Na3 ad 10 Bh3 here, whie 10 Kf2!? ad 11 Kg3!? is aother optio) 10 ... Rc8 11 Be3 cxd4 12 cxd4 Na5 (or 12 ... Bb4+ 13 Kf2) 13 Rc2 Na5 14 Nbd2, D.Navara-A.himaov, Viius (rapid) 2010. 6   cS The positio resembes a fuky Advace Frech, where White gets g2-g4 ad h h4, basicay for free, but whether this represets a sigificat space advatage or a overe xtesio of his positio is yet to be see. 7 Bg2?!

·



A theoretica ovety ad ot a very good oe. Ta's move is ess a pa ad coser to a daydream. A pa requires a strategic or tactica basis ad caot be fabricated  propped up from othigess, from imagatio aoe. It's difficut to fathom just what urgecy prompted the bishop to vacate his orma post ad shift over to the demoted square g2. The bishop simpy doest beog o ths diagoa  a Advace Frech-stye positio, sce it hits a soid wa o d5. We sese a ack of cotuity, a breakg away from previous tetio, merey for the sake of dog somethg differet. 7  e6 8 Ne2 : If d5 represents a soid wa as yo say,

 Two reasons:

the why ot chip away at it with 8 c4 - ?

1. It vioates the prcipe: Don't open the position when la ggin g in

development. 2. The move destabies White's etire cetre ad begs overextesio. Back ca cotue vigorousy with 8 . . . Nc6! ad White's cetre crumbes. 8  Bb5! Botvik achieves the goodificatio (it woud be best ot to Googe ths "word !) of his formay bad bishop. 9 Na3?! This most atura of moves may be accurate, as we sha ater discover. 9 dxc5! Bxc5 10 a4! s a superior method of gog after the ight-squared bishop. 9  Bxe2 The first imbaace. Botvik is quick to had over bshop for a kight, to achieve a Frech, sas bad bshop.

10 Qxe2 cxd4 Principle: Rigid structures tend to favour knights  bishops. This aso enates any thoghts of a ater c3c4 on Ta s side.

11 cxd4

 (  ): Wod yo pay 11 ... Bxa3, handng White both bishops

for two kights or is

it better to refrai? We have diametricay opposite choices. Which of the hypotheses yieds the greatest profit for Back?

 Back shod chop the kight.

11 000 Bxa3! The bishop breaks a botte over the kight's head, ametg: "what a waste of exceet scotch! The cocave deveops to a schism of opposg ad atagostic camps. Botvik correcty assesses that his par of kights wi be a equa or superior force to White's two bishops.

: Didn't Back just greaty weaken his dark sqares on his ast move?

 He weakened them a itte and there is a toch of danger nvoved. Bt with his ast move,

1. He set up a hoe o c4 for hs pieces. I particuar, he pas a future ... Nc6-a5-c4, whe his kight domates. 2. He damaged ad weakeed White's queeside paws, hadg him a pair of doubed isoais o the a-fie. Ths may become a factor shoud quees come off the board. 12 bxa3 Nc6 13 Be3!? Back achieved two goas:

: A bod sac of a3, or a presmptos overreach?

 A itte of both! No power short of death mofied Ta's st for war. He occasionay ndged n frezied excess of raw emotion on any given day. It's one thng to sense an awf trth abot the herent defects n your position; it is entirey another to acknowedge this trth. This move may we deserve a "?! mark. Rash decisions made n the heat of passion have a way of deandng sober amendment when cam reason fnay ss n. For most peope, fear of deprivation, even f ufounded, is a pow erf motivatng factor  bt apparenty not to Ta, who offers, and contnes to offer, his a3-pawn, considerng it deadweight, rather than defend awkwardy with 13 Qd3 hS 14 gxhS Nh6 15 Be3 NfS, when White is on the defensive. t, this ooks ke a better bet for White than the game's contnation.

13 000 Qa5+! Iducig ite to foregog castg rights.

14 Kf1 Nge7 Here we see Botvik's aversio to paw grabbg. I woud take o a3 straight away ad ot give Ta a opportuity to cover it.

15 Rb1 Rb8

16 Bh3!? 16 Qd3 just ist  Ta's vocabuary. 16 000 Qa4!? 17 Rdl!? Last chace for 17 Qd3.

: Don't free pawns atter to either payer?

 Apparenty both show n amost sprita contempt for a thngs earthy. I'm with yo nd wod snatch the pawn at the very frst opportuty.

17 000 Qxa3 It aost fee ike Botvik accepts the offer grudggy, as if dog Ta a favour. 18  Qa6! 19 Qxa6!? I'm certa Ta didt make this decsio ighty, eterg a edg a paw dow agast a superior tecica payer.

: Why not pay the qeen over to d2 or b2?

 Ta undobtedy feared oss of tempi after ... Na5-c4. 19 Qc2 is the ony way to keep qeens on the board withot droppng another pawn, bt then Ta probaby didn't ke the qeen sittng on the open c-fie, where it is sure to ose a tempo to a fture ... ReS.

19 000 bxa6 Back's extra paw may be doubed ad isoated, but it's sti extra!  h5!

So that Back cat seak  . . . h6-h5 at a opportue momet. 20  Kd7 Now it's safe for the back kg to come out of hidg.

21 Rbl Rb6





The defece forms a tight crce which ite is uabe to pierce. Botvik gives hs oppoet a choice of fixg Back's structure or osg cotro over the oy true ope fie o the board.

22 Kg3 Na5 Heo c4-square. 23 Rxb6?! Too much of a cocessio. Ta had to gr ad bear it with 23 Rhcl Nc4 24 Bfl Rc8, whe hs positio s admittedy sti rather poor. 23   axb6 24 £4 Nc4 The kight cotues to swim aps with udimished vitaity.

25 Bl ite's bishops, oce thought so formidabe, have bee tamed ad brought ow.

: White's ast move ooked passive. Why didn't Ta pay 25 Bf2 nstead?

 Ta ooked to ftre f4-f5-f6 tricks, when a bishop on the c1-h6 diagona is st aimed at the h6pawn.

25  Nc6 The kights set aside serviity ad go to war, chaegg their couterpart bshops with the amost sutg sef-assurace of those who eter batte aready kowg of the outcome.

26 Rdl

 (  ): As mentioned before, Ta pans f4-f5-f6 to underne

Back's cotro over h6.

This probem woud be eimated if Back forced a swap of White's dark-squared bishop. How did Botvik accompish ths feat?

 tep 1: Gan a tempo on the white a-pawn.

26   Nb4 27 a3 Step 2: Corer the dark-squared bishop. 27  Na2! The overs redezvous  the mooight. The kights wader where they wi, causg dsruptio with cotemptuous disregard for the aw which states that bshops are their socia superiors.

28 £5 Nxc1 29 Rxcl b5 29 ... Nxa3?! doest make much sese sce White regas the paw with a touch of piece activity after 30 Ra1.

30 Ral Ke7 Botvik cotues his vigiace over f5-f6 tricks, which represet White's oy chace of a swde  this game. 31 Kf4 Rc 32 g5 hxg5+ 33 Kxg5 Itedg f5-f6+ . 33  exf5!

No more f5-f6 worries. 34 Bxf5 Rc6! The third rak proves to be the best postig: 1. Botvk thiks about . . . h6 ad possiby future . . . f7-f6+ ideas. 2. Back's rook may back up a future queeside passer o b6 or a6, after activatio of his queeside paw majority with a evetua ... a6-a5.

35 Kf4 R6 36 Bg4 Rc6 Botviik stas, udoubtedy to reach the time cotro. 37 Rcl f6!? Hdini frowns on this move. Another possibity is to eave the kngside aone and psh his ajority forward with 37 ... a!.

38 Bf5?

 ( ): Ta, by now dejected, bunders

see Botvik's trick here?

 38  e5+ 39 xe5 Nxe5!

away paw umber two. Do you

The armoured kight raises a metaic fst ad smites e5. Apparety Back's hagig kight is ot so hagg.

40 Rdl : What abot pUng the kight?

 40 Re Kf6! covers the kight, snce 41 xe5?? is met by 41 . . . Rc4+ 42 Be4 xe4+ 43 xe4 dxe4 44 Kxe4 (the white kng smes weaky nd asks my friend, my od friend, srey the time for vengeance nd petty recr mnation is ong past?  obviosy Back's kng, who contnes to nrse a grdge, strongy disagrees) 44 ... Kg5 and White's odds fa to zero.

40  Kd6 41 Be4 RcS 0-1 Two cean pawns down aganst Botv is two too mny! Game 37 AMea Gaa-MBotvk Paa de Maorca 1967 Pirc Defence  e4 6

Late  his career, Botvk dabbed  o-cassica opeg es, such as the Prc. 2 d4 Nf6 3 Nc3 g6 4 Nf3 I those days most payers payed the od cassica e, ow reativey ufashioabe compared with the 4 f4 Austria Attack ad the more fexibe 4 Be3 variatios. 4  Bg7 S Bc4 I fd the Bc4 es effective agast the Prc or Moder Defeces. The move s more of a affectatio of aggressio by the socia-cimbg bishop, rather tha ay rea damagg bow to Back. S  c6 The most precse move order.

6 Bb3 This is the ast time ths bishop s dested to move  the etre game. The treacherous piece decides his ma oyaty is to himsef, rather tha his kg. The drect 6 e5 achieves itte after 6 .. . dxe5 7 Nxe5 0-0 8 0-0 Nbd7, whie 6 a4 rus to 6 . . . d5! ad Back has equaity at mmum. 6  0-0 7 Qe2 : What is the idea behnd this move?

 In any Bc4 nes aganst the Prc, White hopes to disrpt by ramng his e-pawn nto Back's position by e4-e5 and sometimes e5-e6. The qeen move backs p this pan.

7  Bg4! 8 h3 Bxf3 9 Qxf3 e6!!





A radicay ew strategic cocept at the time.

: This move ooks unatura. What is Back's idea?

 Havng given p his ight-sqare d bishop, Botvk switches his strcture to a c6 e6 d5 foration, favourng his reanng bishop and shttng ot White's b3-counterpart. In my book  . . . d: Me  Me, this concept is sed by Back over and over agan, and payers who tze this system owe Bo tvk a round of thaks for comng p with the idea.

10 Bg5 h6 11 Bh4 Nbd7 12 O-O-O?! After this atura move Back's attack arrives quicky. Perhaps White shoud be ess ambitious ad sed his kig the other way with 12 0-0 with approximatey eve chaces, W. Watso-W.wic, Jeeia Gora 1980. 12  Qa5 13 Qe2 b5 14 £4 This move weakes e4 but I'm ot sure what ese to suggest. White must do some thg or risk aiatio o the queeside. 14  b4! 15 Nbl d5!





Stressg the efficiecy of White's bshop, who is determed to keep vigi o b3. Vigi for what exacty, o oe ca say. 16 Nd2 cS! The c-fie opes. Back's attack arrives with aarmg rapidity.

17 Bxf6 : Why give p the bishop par when he can gan a tempo with 17 e5 - ?

 The trobe is after 17 ... Nh5! Back threatens both ... Nxf4 and ... c5-c4, smotherng the fortnate b3-bishop.

Istead, 17 exd5! keeps White's disadvatage to a mimum by foowg the prcipe: Counter in the centre when challen g ed on the win g. For exampe: 17 . . . exd5 (or 17 ... Nxd5 18 f5!) 18 Bxf6 Nxf6 19 dxc5 Qxc5 20 Qd3 Rfe8 (toyg with ... Re3) 21 Rhel Rxel 22 Rxel a 23 Ba4. 17  Nxf6 18 dxc5?! To give his bshop breathg room. 18 e5? Nh! s the same story as the previous ote. Perhaps White ca try a desperate couterattack, startg with 18 exd5 Nxd5! 19 f5! gxf5 20 g4 f4 21 g5! Ne3, whe Back's prospects st ook better, but at east White is attackg. 18   dxe4 19 Nxe4 Nxe4 20 Qxe4 Qxc5

: How wod yo rate the chnces?

 White's position is on the critica ist. Athogh thngs seem fne when viewed from a qick

1. Opposite-cooured bshops favour the attacker  ceary Back  this cas e. White's out-of-pay bishop ervousy deces participatio, caimg fatigue ad the eed to beg vespers prayers. 2. White must watch out for ... a7-a5-a4 tricks. 3. White must dea with a back quee trasfer to the al-h8 diagoa, for exampe with . . . Qe7-f6. Cocusio: White's positio is bordere busted, if ot aready there. 21 Rd7?! Beieve it or ot, ths move merey oses time. 21  Rad8! Tacticay aert pay. Botvk avoids the trap 21 ... a? 22 Bxe6! whe Back ca't accept the sac. Nor ca he p the bshop with 22 . . . Rae8?? as that waks to the secodary shot 23 Bxf7+ ! . otward configration, when examned deeper his position is fond ackng:

22 Rhdl? The rather sorry-ookg 22 Rddl was ecessary. 22 000 Rxd7! 23 Rxd7 I chess, as  ife, sorrow aways has a mea-spirited way of arrivg  waves. Now we get to see why opposite-cooured bishops favour the side with the itiative ad attack.

 ( ): Back to pay and w materia.  23



Qg +! 24 K2

The kg awakes, shakg  hs sweat-soaked bed, recag a ightmare too terribe to be put to words. 24 Rdl?? oses staty to 24 . . . Bxb2+ . 24 000 Qf2+! Computer-ike accuracy. The quees face is a portrait of fae cocer. She aows a actress-practiced tear to stream dow her eft cheek, as she proouces to White's kg: "Your Grace, my prayers are for your swift recovery. Botvik's move s much stroger tha 24 ... Bxb2 25 Ke2! Bc3 26 Rdl, whe White cotues to resst.

25 Kd3

The surviva stct s perhaps a eve more eemeta reaity tha the wi to w. White's terrified kg, sesg a ficker of motio to his right, sidesteps potetiay fearfu bows with ches to spare, ad  dog so fas to aeviate his desperate situatio a iota. White's disma ateratives: a) 25 Qe2 Qxf4+. b) 25 Kdl Qf1+ 26 Qel (26 Kd2? Bxb2 threates . . . Bcl mate) 26 . . . Qxf4. c) 25 Kcl Qf1 + 26 Kd2 (or 26 Rdl Bxb2+) 26 . . . Bxb2 is the same story. 25  Qfl + Back's quee appies checks ike war pat.

26 Qe2 Aga 26 Kd2 Bxb2 threates . . . Bcl mate. 26   Qxf4 Botvik's imby destructive quee ficts damage with bdg speed, the daces away to orgaize yet aother bow.

27 Qf3 27 Rxa7?? Qd4 mate woud be somewhat coveiet. From ths pot White's resistace sks from view, the way risg foodwaters overtake ad eveop a sma tow. Note how White's useess bishop, ot deigg to participate, merey observes the proceedgs, eyebrows raised  disda, from a digified dstace. Just compare the power of the bishops o the board. So margaized is the pathetic tw o b3, compared to his g7-brother, that the par whe added up come out to a umber ess tha two. 27   Qe5 28 Qb7 as 29 Qc6 Qxb2 The creature uges, its terribe madibes cickg ad sappg  aticipatio of the fesh it w soo taste. Ceary, there s o savg White, ow that a ve has bee opeed o b2, freeg c3 as yet aother probem square for White. 30 Ke2 Qe5+ 31 K h5 32 R7 Qf4+ 33 Ke2 B4! 34 R Qe3+ 35 K1





I the midde ages a pubic beheadg was cosidered a festive ad ceremoia evet  whoesome fu for the etire famiy ad a peasat way to sped a Suday afteroo. The path to victory grows trasucety cear.

 ( ): Back to pay and force mate.

 Dobe attack: mates are threatened on e and d2. The white kng's crses are wasted on Back's bishop, who merey appears amsed.

35  Bc3! The kg drops from the gibbet, his egs kickg as he sways. With the witessg crowd, his coivg bshop ad quee, brother ad sister, whisper: "Let us ever utter his vie ame aga. 36 Rxf8+ Kxf8 0-1 Game 38 MMatulovc-MBotvk Word vs. R match, Begrade 1970 Pirc Defence

Compare B otvk's pay n this game to the previos one. They are eeriy siiar.

1 e4 g6 2 d4 Bg7 3 Nf3 d6 4 Bc4 As I metioed ast game, o c4 the bishop is a skied professioa  the wrog professio, with o pace to appy his ski. I my opio, the Bc4 es, especiay agast the Moder Defece, fai to fit the overa pa, ike mismatchg furiture which sours the decor of a otherwise eegat room. I The Modern Defence Move by Move I rather rudey ca ths e the Cro-Mago Variatio. 4  Nf6 Trasposg from Moder to Prc. I my book I advocate a wa, erected with 4 ... e6, to but the bishop's fuece, whe it ca oy hobbe aroud with impged fuctio o the a2-g8 diagoa.

5 Qe2 Just sighty aterg from the previous game. Qe2 is give precedece over Nc3. 5  c6

6 Bb3 : hodn't White srge forward with 6 e5 this time?

 I beieve this is his oy chance at an edge. For exampe: 6 ... Nd5 7 0-0 0-0 8 h3 dxe5 9 dxe5 (after 9 Nxe5 Nb6! 10 Bb3, J.Mon-L.Fressnet, French Championship, Pa 2012, Back may be abe to get away with grabbng the pawn; i.e. 10 ... Qxd4 11 Nf3 Qd7 12 Rd Qc7, when I don't beieve n White's compensation) 9 ... Qc7 10 Re h6 11 Nbd2 and White's extra centra space and potentia kngside attackng

chances ay give him an edge, V.Krak-R.Ponoariov, Moscow (bitz) 2008.

6  0-0 7 BgS Matuovic's attempted mprovemet over Botvk's game with Meda.

: Bt isn't this the same position?

 Not qite. White eft ot Nc3 and repaced it with an earier Bg5.

: What is the signficance?

 It means that Botvik does't have access to his (then) revotionary ... Bg4 pan, any becase White responds with e4-e5 as soon as Back pays ... e7-e6. However, Matovic soon earns that it is not so easy to otprep Botvk, who is ready for Matovic's idea as we.

7  h6!

: Why an excamation mark for sch a mundane move?

 Pease ask me this qestion agan on move 11! Perhaps Matovic expected Botv to contne n siiar fashion to the previos game with 7 ... Bg4, when 8 Nbd2! gives White a sight advantage.

 Bh4 eS!





Botvik itiates a radica ew reform to od party doctre by switchg gears ad settg up a dark square cofiguratio stead.

9 dxe5 dxe5 10 Nbd2 10 Nxe5?? is deat with harshy by 10 ... Qa5+, pickg up the kight. 10  Qc7! Aga, amazg precisio. Botvik ups rather tha deveops. 11 Nc4 NhS! Now we see the idea behd Botvik's 8 ad 10 moves: ... Nf4 becomes a

oyig issue for White. 12 Bg3 Nf4! Pickg up the bishop pair ad the dark squares with it. The obvious 12 . . . Bg4? is a mistake. White grabs a advatage after 13 Bxe5! Bxe5 14 Ncxe5 ad  14 ... Nf4 15 Qd2! Nxg2+ 16 Kfl Bxf3 17 Nxg6!, his positio borders o wg. 13 Bxf4 exf4 14 O-O-O?! A few gras of sad escoced withi the works of a watch are eough to throw off the fey tued gears. White reaches the high pot of hs "itiative. His positio is dested to degeerate  ati-cimatic fashio. GM Aex Vozhi writes: "A very risky approach. White makes the same mstake as  the game Meda Botvik, ceary uderestimatg the power of Back's dark-squared bishop.





Back aso stads better if White refras from castg ad pays 14 e5 Bg4 15 Qe4 Nd7 16 Qxf4 Bxf3 17 Qxf3 b5 18 Ne3 Nxe5 19 Qe2, but perhaps ths s how White shoud have payed to mimze his disadvatage. 14 000 Bg4 Aimg for opposite-cooured bishops oce aga.

15 e5 Or 15 h3 Bxf3 16 Qxf3 Nd7, whe . . . b7-b5 s  the air, ad  17 Nd6 Ne5 18 Qxf4??, the 18 ... Rfd8! ws a piece: the white kight caot move because of ... Nd3+ !, wg the quee, whie 19 Qd2 fais to 19 . . . Bf8. 15 000 N d7 16 Qe4 Rad8 Botvik sorts at the offered trife. Of course he is ot about to squader  good positio for a crumb after 16 ... Bxf3 17 gxf3 Nxe5?! 18 Nxe5 Qxe5 19 Qxe5 Bxe5 20 Rd7, whe White IS cotro of the seveth rak compesates the missg paw. 17 Qxf4 Bxf3 18 Qxf3 bS 19 Ne3 NxeS 20 Qg3 as

·



reateg ... a5-a4. Havet we see this movie before? As  the previous game: 1. Back ows the stroger bshop. A i geera's rak is subordate to the wi of hs ower rakg army doctor, sce the geera's body s actuay the property of the goveet, ot his ow to degeerate wifuy. hite's bishop, which ooks powerfuy posted o b3,  reaity does itte ad is merey a statioary target for ... a5-a4 or ... c6-c5-c4 paw thrusts. 2. Back ejoys domatio of the dark squares . 3. Back has the stroger attack.

21 a3 K7 Back might ao cosider givg up the d5-square to auch hs armada straight away with 21 . . . c5! ? 22 Bd5 c4.

22 Ng4 5 23 Nxe5 Bxe5 Oce aga, Botvk's bishop towers over  couterpart o b3.

24 Qf3 Kg7 25 Re1 Bf6

: I don't reay see Back's advntage

or attack. Ist the positio rather drawish?

 Patience nd a ufods. hite may even be bsted here. Once agan, the mbaanceprncipe: Opposite-coloured bishops four the attacking side  and generate terribe force n jst a few moves.

26 c3 : O his ast move, Matovic weakened his kng position. The

questio  how woud

Botvik attack if hite sat tight ad did othig?

 A sampe ne: 26 Qg3 (the nervos qeen ss her teeth nto her ower ip) 26 ... Qb7! (of course Back is unwng to swap) 27 Kb1 xd1 + 28 Rxd1 a4! 29 Ba2 b4 (targetng b2) 30 axb4 Qxb4 31 c3 Qe4+! 32 Qd3 (32 Ka1? a3 33 bxa3 Qc2 eads to ate) 32 ... Qxg2 and hite is bsted, down a pawn, perhaps osng more, and remang with an exposed kng.

26  cS Here they come. 27 Rxd8 Rxd8 28 g3 c4 29 Bc2 A sodier returg home from a faied, upopuar war shoudt expect a victory parade. 29  b4!





Botvik reetessy goes after ite's kig. The famed sore o b2 begs to throb to a ubearabe eve. 30 axb4 axb4 31 cxb4 Qb6 32 Rd1 Qxb4?! This  too eemeta a reactio for a positio requg fiesse ad was udoubtedy a hasty decisio, fueced by the cock. Botvik jeopardizes his w by aowg too may swaps. He shoud reta a pair of rooks o the board with 32 .. . Rb8!. 33 Rxd8 Qxb2+ 34 Kd1 Bxd8 35 Qd5 Qf6 The quee returs home to cout her moey.

36 Ke2 Bb6 Forcg a further weakess.

37 f4 Qc3

38 Qe4?! If you sese the oppoet's mometum gatherig stregth, sometimes your best bet is to cofrot the meace immediatey, rather tha sta a fight that  ievitabe. Here 38 Qe5+! Qxe5+ (the quee petuaty agrees to the egotiatio, may sice she had o choice  the matter; ow ite's harried kg begs to reguate his oce ragged breath ad Back's attack gutters out, ike a cade ear a ope wdow) 39 fxe5 Bd4 40 e6! does't ook so easy for Back to w, despite  two extra paws. 38  Bd4! Not givg ite a secod chace to remove quees from the board. 39 f5? By attemptg to expose Botvk's kg, Matuovic exposes his ow. After this move, his ow foudatio sags ad coapses. 39  gxf5! Defectio. 40 Qxf5 Qe3+ Back's quee eters with deady effect.

41 Kf1 Not 41 Kd1?? Bc3! etc. The bishop is one of those scheng peope who tends to enter rooms unoticed and unseen. This is the identica matng pattern from Medna-Botv, a game Matovic mst have stdied heaviy n his pre-game prep!

41  Qg1 + 42 Ke2 Qxh2+ 43 Kd1 Qg1 + 44 Ke2 Qe3+ 45 Kf1 Qxg3

·



te's quee cat take aythg sce she must stad guard agast mate o £2. Nor does a perpetua check exst aywhere. 46 Qh7+ Kf8 0-1 The back kng's person is every bit as sacrosanct as the popuace beieves. The cur of his ip (an ufortunate genetic trait herited from a great aunt) makes him appear 35% more arrognt thn he actay is. After 47 Qf5 Qg+ 48 Ke2 Qg2+ 49 Kd B c3! we veer coser to that famiar pattern once agan!

Chapter Five Botvinnik on Accumulating Advantages A over-the-hi major eague baseba payer may sti shie if demoted to the mior eagues. Botvik i hs game agast Sziagyi (Game 46), may have bee past hs prime but, whe facig a o-word-cass payer, sti packed a puch. Ths game is oe of the cearest exampes of seizig cotro over ad the expoitg a weakess o oe coour. If we payed a chess word association game and said " Accmatng Advantages, I wod repy "Botvk ! The power of Botv' s ogic endures and the once fragmented segments coaesce nto a perfect whoe. His microscopicay detaed strategic surveince reveaed the most hidden nances, unseen by the comon eye of the majority of his opponents and a strategic advantage, once sezed, was rarey renqished. Game 39 BSF Moscow 16 Alekhine'5 Defence  e4 Nf6

Aekhe's Defece was a rare guest  the 1930s. I my database, I cout oy 19 times whe eve Aekhe himsef payed hs ow Defece, ad a but two of these were  the previous decade . 2 e5 Nd5 3 d4 d6 4 Nf3 The ma e, perhaps hite's best shot at a edge . 4  Bg4 Currety, most top GMs favour the e 4 ... dxe5 5 Nxe5 c6.

5 Be2 c6

6 0-0

Sometimes ite goes for 6 Ng5 ad I aways respod with 6 ... Bf5 7 e6 (amost everyoe pays ths, but I'm covced 7 Bd3 s a better try for a edge) 7 . . . fxe6 8 g4!? (after 8 0-0, I hug o to my extra paw ike grim death ad wo  N.De

Frmia-C.Lakdawaa, Word Rapid Champioship, Mazata 1988) 8 ... Bg6 9 Nxe6 Qd7 10 Nf4 Nxf4 11 Bxf4 e5! (a theoretica ovety) 12 dxe5 dxe5 13 Bxe5 Qe6!? with . . . Nd7 ad 0-0-0 to foow, with a ferocious itiative for the paw, A. HacuC. Lakdawaa, Sa Diego 1984. 6  dxeS This eads to a passive positio. Today, most payers prefer 6 ... Bxf3 7 Bxf3 dxe5 8 dxe5 e6 9 Nd2! (beieve it or ot, the edig s quite okay for Back after 9 c4 Ne7 10 Qxd8+ Kxd8, sice the e5-paw provides a target for Back which easiy offsets White's extra space) 9 ... Nd7 10 Re Qc7 11 Nc4 N7b6 (11 ... b5? 12 Bxd5! cxd5 13 Nd6+ ! Bxd6 14 Qxd5!, aother dirty trick worked out by the computers, s very good for White) 12 Qe2 Nxc4 13 Qxc4 0-0-0, as  A.Ivaov-C.Lakdawaa, Los Agees (rapid) 2000. Back's positio is payabe though aga sighty passive. I wet o to get squeezed by my oppoet's space ad bshop pair. 7 NxeS Bxe2 8 Qxe2 Nd7 9 f4 Rug out . . . Nxe5 possibiities, sce the recapture with the f-paw opes the f fie for White's rook. Aterativey, V. Maakhov-D. Labuskiy, Russia Team Champioship 2001, cotued 9 c4 Nxe5 10 dxe5 Nc7 11 Nc3 g6 (11 ... e6? eaves a gapg hoe o d6) 12 Be3 Bg7 13 f4, whe Back foud himsef cramped ad short o couterpay. 9  e6 10 c4 NSb6 11 Be3 Back may be passive ad cramped, but it ist easy to expoit, sce he remas free of weakesses.

11   Be7 12 Nc3 0-0 13 Rf3!? The youger Botvik aways had oe eye o the oppoet's kg. Most payers woud opt for the more orma 13 Rad1. 13  Qe8!?

: What is the pont o f this move?

 I thnk Back was worried abot  and Qh5, so he paned . . . f7f5 at an appropriate moment.

14 Rdl Rd8 1S b3 fS! Back's kg ooks a bit safer after ths move.

: Bt doesn't the move aso create a backward e6-pawn?

 As it trns ot, e6 isn't so easy to expoit snce Back easiy defends the pawn as any tmes as White attacks it.

16 Nd3 Principle: The side with the extra space should retain pieces on the board. Botv tries to ake somethng from his space edge bt has trobe akng headway. Back fees apprehension bt not yet fear.

16  Bf6 Back couterattacks d4 to offset White's potetia pressure o e6.

17 Bf2 Qf7 18 Ne1 Itedig Rfd3 ad Nf3. 18   Rfe8 19 Rfd3 Nf8 20 Nf3 Qc7 21 Ne5 Nbd7 22 Qd2 Botvik decides that e6  a uattaiabe target ad aows Back to sea the square with ... Nxe5.

22  Be7 22 . . . Nxe5!? 23 fxe5 (23 dxe5 offers othg after 23 . . . Rxd3 24 Qxd3 Be7, as White has o fitratio pots dow the d-fie  Back ca cover a7 ad the chaege with a future ... Rd8) 23 ... Be7 is aso payabe. It wot be so easy for White to try ad expoit his extra space sce Back acks eve a sge weakess  his positio. 23 Nf3 Nf6 24 Qcl Ne4! 25 Ne5 25 Be3 Bb4 is aoyg for White. 25  Nxf2 26 Kxf2 N d7 27 Qe3 Nxe5 28 fxe5 28 dxe5 Rxd3 29 Rxd3 b6 30 Ke2 Rd8 ooks equa. 28  Qa5 29 a4 Rd7 30 g3 Qd8

31 Kg2 : Can White break throgh with 31 d5 - ?

 If Back is caref he cn hod his own after 31 ... exd5 32 cxd5 Qa5! (not 32 ... cxd5?! 33 Nxd5 with a cear advantage to White, who threatens Nf6+ as we as Qxa7) 33 Kg2 (33 d6? overextends after 33 . . . Bd8 !, threateng to swng round to b6, and i 34 b4 Qxe5 35 Qxa7 Bg5, Back is better snce White's kng cover has been eroded) 33 ... Bb4 34 dxc6 xd3 35 Qxd3 bxc6!, when Back ooks fne.

31 000 BgS 32 Qf3 Qe7 33 cS!

: Why did White jst give hisef a

backward d-paw rather tha pay d4-d5 - ?

 Botv didn't want to reease the centra tension too eary, and paned b3-b4-b5. Let's ook at your sggestion: 33 d5 exd5 34 h4! Bh6 35 cxd5 Qxe5 36 dxc6 xd3 37 xd3 bxc6 38 Qxc6 Qe6 39 Qxe6+ xe6 40 Rd7 Re3 41 Nd5 xb3 42 Rxa7 Be3 and agan Back shod hod the position.

33 000 as!

 (  ): With his ast move, Fohr stopped the

b3-b4-b5 pa. White must ow

switch gears ad fid a aterative idea to make progress. Come up with a pa for White.

 34 Nb!

Botvik pas to trasfer  kight to the domiat d6-square via a3 ad c4. 34  Qf8? Fohr drifts  i fact after this move his positio ooks strategicay ost. Back had to pay 34 . . . Rd5! 35 Na3 b6! 36 Nc4 bxc5 37 h4 Bh6 38 Nxa5 Qc7 39 Nc4 cxd4 40 Rxd4 g6, whe he cotiues to fight o.

35 Na3 Bd8 This time the . . . b7-b6 trick fai mseraby after 35 . . . b6?? 36 Qxc6 ad White W.

36 Nc4 Bc7 37 Nd6 The eve of offece is heighteed by the simpe fact of the rude kight's absoute ack of cotritio. 37  Rb8





What a differece from a few moves ago! If Back chops that d6-kight, he hads White a deepy embedded passed paw ad opes the e-fie as we, which edagers e6.

 (  ): Fnd a new pan for White to ake progress.

 Open the b-fie and tar get b7.

38 Rbl! Qd8 39 b4 axb4 40 Rxb4 White's aim is to force Back to take o d6. Back must grudgigy sette for this ufortuate exchage sice the pressure o b7 grows to ubearabe eves. 40  Bxd6 The bishop's upper-bracket come takes a tumbe, as he oses his od job ad embarks o hs ew career as a street mime. Back reuctaty decides upo a

remedia surgica optio to remove the famed sore from d6, oy to have it repaced with aother famed sore o d6!. Strategic suicide, but it's uderstadabe that Fohr didt wat to edure a sow death after 40 . . . Ba5 (the bshop fiay reaches his destiatio, oy to wish he were eewhere) 41 Rb2 h6 42 Rdb3 Qe7 43 h4 (Back is cose to zugzwag) 43 ... Kh8 44 Qd3! with Qb1! to foow.

41 exd6





 upeasat protuberace pops up o d6, which Back fids as difficut to

remove as a wart. New targets arse at e6 ad b7. hite's prste structure shies  comparso to Back's diapidatios, which spraw across the board. 41 000 Qa5 42 Rdb3 Re8 43 Qe2 Botvik pays it safe, deyg Back ay potetia couterpay after 43 Rxb7!, which Houdini works out to a w for hite  the sequece 43 . . . Rxb7 44 Rxb7 Qxa4 45 Qh5 Rf8 46 Qe2! (ow e6 fas) 46 ... Qxd4 47 Qxe6+ Kh8 48 Rf7 Rb8 49 Rxf5. 43 000 Qa8 44 Re3 Kf7 45 Qc4 45 Qd3 is perhaps more accurate ad avoids Back's ext move. 45 000 b5!?





The desire to retaiate agast a oppressor may be a cosmic uiversa. Back's cever, if sufficiet, idea smacks of a artificia costruct, sce the termissio of Back's pa s of brief duratio. Sti, whe busted, what ee to do but ook for tricks?

46 Qc2 Botvik deces, hurriedy egagg  ateratios, whie dodgg a pair of traps with poite disterest. Back's pot: a) 46 axb5?? cxb5+ ws the quee. b) 46 cxb6? c5+ 47 b7 Rxb7 48 Rxb7+ Qxb7+ 49 Rf3 cxd4 50 Qxd4 Rd8, whe Back may hod the game. 46  Rxd6!? The crossbow was a deady weapo but, oce fred, took a eterity to crak up ad reoad. More tricks, but Botvik has everythg uder cotro. 46 ... Qa5 ad ow the startg 47 d5!! is hopeess for Back as we. 47 cxd6! I a chess game it is impossibe to be ready for every cotgecy, sce we have o cue of ther shapes or forms, or eve if ad whe they arrive at a. Sometimes the oy way to refute a oppoet's trap is to fa to it! 47  c5+ 48 Kh3 cxb4 49 Qc7+ Kg8 50 d7 This paw parayses Back, foatg  rarefied air o d7. 50   Rf8 51 Qd6! h6 Not 51 ... bxa4?? 52 Qxe6+ Kh8 53 Qe8!, whe Back has o defece to Qxa8 foowed by Re8+. 52 Qxe6+ Kh7 53 Qe8!





The quee struts about with the ucoscious ease provided by a ife of priviege . 53  b3! If a ifeguard sees two peope drowg simutaeousy, shoud't he at east try ad save oe of them, eve if he cat save both? So perious is Back's positio, that each move  a matter of grave faity. Fohr fds Back's best chace. Istead, 53 ... bxa4? 54 Qxa8 Rxa8 55 Re8 eds it, or 53 ... Rxe8? 54 dxe8Q Qxe8 55 Rxe8 bxa4 56 Ra8 ad Back's paws prove to be too sow  ther dream to promote.

54 Qxa8 Botvik simpifies to a easiy wo rook edg. 54  Rxa8 55 axb5 Rd8 56 Rxb3 Rxd7 57 b6 Pnciple Passed pawns must be pushed.

57  Rb7 For a gravey wouded sodier, the mere fact that he remas amog the ivg  ot ecessariy a cause for ceebratio. Back's rook is hopeessy tied dow. 58 Kg2 Kg6 59 Kf3 Kf6 60 Rb5 Ke6 61 Ke3 Kd6 62 Kd3! Kc6 63 Kc4! Kd6

63 ... Rxb6 64 Rxb6+ Kxb6 65 Kd5 g5 66 Ke5 is game over. The back kg's eyes are poos of hopeess resigatio to a terribe fate.

 (  ): No need for a convoted, rambng pan.

Fd oe straightforward idea ad you force Back's resigatio. How did Botvk covert hs advatage  this titaic strugge?

 Cear b5 for his kng.

64 Rd5+! Kc6 65 Rc5+! Kd6 Back's kg, who aways regarded himsef as head of the ca, the paterfamiias, ow reaes he was mistake, sce ite's kg seems to have assumed the roe. 65 ... Kxb6 66 Rb5+ simpy trasposes to a dead ost kg ad paw edg.

66 Kb l-0 Is it not treason to obstrct a kng's wishes? Game 40 VSmyslov-MBotvk Word Championship (15th atchgame), Moscow 1954 icilian Dence  e4 c5 2 Nc3 Nc6 3 g3 g6 4 Bg2 Bg7 5 3 6 6 Nge2

Today, theory says pay the kight to f3 (after f2-f4) or ee to h3. But either pa offers ite a edge. 6 000 e5!!

: A dobe excamation mark for a rotne move n the Cosed iciin?

 It's a dobe excam i yo nvented the entre idea, which bossomed nto two ajor branches of both the iciian and the Engish (when payed a tempo p). This was the very first tme this idea was payed, and it had the effect of absotey deciatng mysov's home prep. In fact, he abndoned Cosed ician ician after afte r this game.

The remarkabe remarkabe strategic stra tegic cocept behd Botvik's Botvik's move s that Back ca de deed ed get away with the creatio of a giat hoe o d5, due to hs ampe compesatio of cetra cotro. I fact, it becomes apparet that ite's cotro over d5 gives him itte. Back's ast move must have come as a compete shock to the witessg audiece, ad Botvk Botvk may have bee a bit ike a frog tryg to expa the cocept of breathabe breathabe air to fsh  smpy outside the ream re am of comprehesio. comprehesio.

7 Nd5 Nd5 Smysov, udoubtedy started by Botvik's ast move, jumps  kight to the hoe, probaby expect expe ctg g a a edge . 7 0-0 ad ad 7 Be3 Be 3 are are payed paye d more ofte today. 7   Nge7 Nge7 8 c3 c3 I dot thk it's a wse idea to aow Back to sea the d5-square, ite's sge pus here. Smysov probaby saw that ateratives ead to ess dyamic positios: a) 8 Bg5 h6 h6 9 Bf6 0-0 10 Nec3 Bxf6 11 Nxf6+ Kg7 12 Nfd5 Nxd5 Nxd5 13 Nxd5 ad hite got othig, Bpassky-L.Chrstiase, Lares 1985. b) 8 Nec3 0-0 9 h4!? (the push of hite's h-paw s aother try  this e) 9 ... Nxd5 Nxd5 1 0 Nxd5 Nxd5 Ne7 1 1 Nxe7+ Qxe7 12 h5 h5 Be6 Be 6 13 1 3 Be3 b5 14 Qd2 a 15 a4 b4 16 1 6 Bg5 (a waste to time sce Back Back wat watss to pay pay . . . f7-f6 f7-f6 ayway ayway)) 16 . . . f6 17 Be3 Rad8 18 hxg6 hxg6 hxg6 ad Back stads sighty better with  greater space, whereas hite's kgside attack is oexistet, Y.Baashov-V.Fiippov, Kaza 2005. 8  Nxd Nxd5 Botvik f the d5 hoe with a white paw. 9 exd exd55 Ne7 10 0-0 0-0 0-0 0-0 11 f4?! f4?!

·



What was oce a atura fow ow aters to somethg viscous ad syrupy. Botvik writes: "After ths pseudo-active move, White gets a difficut game. The paw o [f4] merey imits the mobity of his kight ad QB. The white kg positio is somewhat weakeed as we. I add: whe Back exchages o f4, he gets a heathy kgside paw majority, which coud troube White  a edg. Furthermore, Whi White te weake we akes s e3, e 3, a factor o whic which h Botv Bo tvik ik soo pouces pouces . 11   Bd7 12 h3 h3 Qc7 Qc7 13 Be3 White shoud probaby reease the tesio ad go for 13 fxe5 Bxe5 14 Bh6 Rfe8 15 g4. 13   Rae8 14 Qd2?! Qd2?! This competey atura move ater waks to a uusua p. Botvik suggests 14 Bf2, whe Back sti sti ho hods ds a edge after afte r 14 . . . exf4 exf4 15 1 5 Nxf4 Nxf4 Nf5 due to hs heat heathy hy kgside paw majority. 14  Nf5 Nf5! Botvik uderstads that he ca utize the f5-square. Hs move IS much stroger tha tha the the route route 14 . . . f5. f 5.

15 Bf2 The bishop raises a had, sigag cam to hs comrades, who grow agitated by the back kight's presece o f5. 15   h5! 16 Rael : Wasn't Back's ast move risky

sce sce it aowed aowed White White to surge with with 16 g4 - ?

 This woud qucky ead to White gettng overextended: 16 ... hxg4 17 hxg4 Nh6! 18 g5 Ng4 19 Bg3 c4! with a cear advantage to Back.

16  Qd8! Qd8! Subte positioa sight. Botvk Botvk reaizes he eeds hs quee for the fight o the kgside. 17 Kh2 Bh6!





The od bshop deivers a sermo o the vrtues of respectig your betters  society (i.e (i. e . himsef!) himsef!) . We ca safey deduce that that this this positio was ot what what Smys Smysov ov had  md whe he essa e ssayed yed the Cosed Cose d Sicii Siciia! a! Now . . . h-h h-h44 is  the the air, which which duces White to his ext weakeg move. 18 h4?!  overreactio which oses materia. The dscomfort of the cure exceeds the itia compat. Smysov made atura moves ad fds himsef strategicay busted! Such was the awesome power of Botvik's positioa uderstadg. 18 000 Qf6! 19 Be4? Apparety, oy the werd computer move 19 Rd1! aows White to hag o by a thread.





 the oe side there s hesitat discatio; o the other we fd cofidet preparatio. White fais to appreciate the true import of Back's ast move ad his game sps out of cotro. The tesio reaches its apex ad ow the bu begs to scrape the groud ad sort, preparg the charge.

 ( ): How did Botvk wn ateria n this position?  19



exf4! 20 Nxf4

The kight's kight's soft, utroubed utr oubed hads hads betray be tray a ife of ease ea se ad comfort  but this this is about to chage.

20  Nxh4! Nxh4! Overoad.

21 Be3 Be 3 Nf5 22 Bxf5 Bxf5 A miserabe move to be forced to make. Smysov sees his possiby iteded combatio 22 Nxh5? gxh5 23 Bxf5 Bxe3 24 Rxe3 rus to 24 ... Qg5! 25 Bxd7 (25 Be4?? is met simpy by 25 ... f5) 25 ... Rxe3 26 Rf3 Re with a cea, extra exchage. 22  QxfS 23 Qg2 23 Ne6?? fais mseraby to 23 ... Bxe3 24 Qxe3 Qxd5 25 Nc7 Rxe3 26 Nxd5 Rxe1 27 Rxe1 Kg7 28 Re7 (the fitratio eads owhere) 28 ... Bc6 with two extra paws  Back's accout. 23  Qg4 24 Qe2 Compete desperatio ad the equivaet of resigatio. 24 Bd2 Rxe1 25 Rxe1 g5 26 Ne2 Re8 27 Ng1 Rxe1 28 Bxe1 Bf5 is o better. 24   Qxe2+ 25 Rxe2 ReS 26 Reel After 26 d4?! cxd4!, White woud have to recapture with his paw. 26  Rfe8 27 Bf2 h4!





The demadg demadg taskmaster wrgs wrgs every ouce of sweat swe at from his pieces . After sufferg absoutey zero vicissitudes this game, Botvik ow focuses o coversio. Evets coudt have fowed ad resoved themseves more favouraby if Botvik himsef had orchestrated the story e. We ecouter White's symptoms of decay ad diapidatio across the board: 1. He is a cea paw dow. 2. Back ows the bshop pair. 3. White has a cripped queeside paw majority, which effectivey makes him two paws dow. 4. Back cot cotues ues to chip chip away at the suppor s upportt of f4 with fastidious care . 5. Botvk Botvk creates a par of kgside passed paws. 6. Botvk Botvk pays the back pieces! 28 RxeS RxeS 29 d4 hxg3+ 30 Kxg3 After 30 Bxg3 cxd4 31 cxd4 Re4, Back ws a secod paw. Istead, White's kg, beate dow ad humiiated, raises a fist to o oe  particuar, decarg: "I am resoved ad wi extract revege! Ufortuatey, his predictio fais to come to pass. 30  RgS+ 31 Kh2 RfS! 32 Be3 32 Ng2?? Bb5 eds the game at oce. 32  cxd4! cxd4! Overoadg White's bishop ad hadg him a pair of weakgs o the d-fie.

: Why not wn a piece with 2 .. . g5  ?

 White can pay the trick  Rg1, thogh even here, his position is hopeessy ost.

33 cxd4 Kh7!

Threateg ... g6-g5!.

34 Rf2 g5 35 Ne2 The og-ped kight fiay rouses himse from a state of petrifactio. 35   Rxf2+ Rxf2+ 36 Bxf2 Bxf2 f5 f5 0-1 0-1 By dnt of hard abor and a frga festye, Botv accred a tidy sm for his retirement. Game 4 MBotvk-VSmyslov Word Championship (5th matchgame), Moscow 1957 King's Indian Defence  c4 Nf6 2 Nc3 g6 3 g3 Bg7 4 Bg2 0-0 5 4

Avoidg future d2-d3 Egish set-ups. Botvik teded to foow the phiosophy: if the oppoet o ppoet offers you y ou the cetre, cetre , occupy it with your paw p aws. s. 5  d6 6 Nf3 Bg4!?

To my md this variatio s more dubious tha aggressve, ad perhaps a uwse choice agast Botvik.

: Why unwise?

 In this ne Back prepares to chop the kight kight on f3, handng handng over the t he bishop par. mysov, mysov, ike ike Chigorn before hm, was never shy abot renqishng the bishop par to create an mbaance and pay for the wn. Bt to do so n a sti fid strcture, aganst one of the greatest strategists of a tme, seems overy defiant to me.

7 3 Bxf3 8 Bxf3 N c 6 After 8 ... c6 9 0-0 Nbd7 10 e4 e 11 d a 12 Be3, Back's ack o the ight squares is fet for some time to come, G.Gajewski-Skripio, Iteret (bitz) 2009.

9 Bg2 Nd7 Itesifyg pressure o d4.

10 e3 Much stroger tha 10 d?! which greaty creases the scope of Back's bishop. 10  e5 11 d5 d5 Isurg that Back's remag bishop wi ive hs ife surrouded by paws of the same coour. 11   Ne7 12 e4 e4 £5 £5 13 h4!?

: Is Botv attemptng to aunch a kngside attack?

 Perhaps he considers h4-h5, bt I do't th this is his an motivation. Instead, he pns a fture Bh3!, ncreasng fence over the ight sqares.

13  £4?! A dreadfu sef-ficted woud o the ight squares. Back shoud mimze his

dsadvatage with 13 ... Nf6.

: Why wod a payer as strategicay acte as mysov pay sch a move?

 mysov's idea is to rid hmsef of his own bad bishop with ... Bh6 and eventay swap it for White's White's dark-sqared dark-sqared bishop. bishop. Bt the price of o f ight ight sqare sqare damage exceeds the benefits. benefits.

14 B3

Of course! The bshop's uwated presece presets a impositio o a i Back's camp. 14  Rf6 15 Qe2 Bh6 16 Bd2 ite ca ao go for a drect assaut with 16 h5! (hopg to duce the awfu ookg ... . . g6-g5) 16 . . . Kg7 17 hxg6 hxg6 hxg6 18 f3 f3 c6 19 gxf4 Bxf4 Bxf4 20 Bxf4 Rxf4 21 0-0-0 Nf6 22 Be6 cxd5 23 cxd5 ad Back's kg s  serious troube. 16  Nc5! Nc5!?? Smysov ooks for troube, ducg b2-b4, arguaby a move hite wats to pay.

17 b4 b4 f3 f3 18 Qf1 Qf1 Of course ite must keep vigi over d3. 18   Bxd2+ 19 Kxd2 Kxd2 Na6 Na6 20 a3 c6

: Isn't White's kng n danger?

1. ite's ufchg kg ist as vuerabe as it ooks, sce  massive queeside space advatage offers him ampe protectio. 2. ite's cetra paws are fixed o the same coour as hs bishop, yet his bshop is aythg but bad, due to its scorchg fuece o the h3-c8 diagoa. 3. The eemy f3-paw, for ow the bae of hs existece, is for a tets ad purposes a isoai, sce it acks support from eighbourg e-ad g-paws. This meas Back is ikey to drop the paw shoud more pieces come off the board. Cocusio: hite's umerous strategic advatages easy outweigh the mid dager to  kg. 21 Qd3 Nc7 22 Rab1 Rb8 23 Rhc1 If cetre ad queeside ope too quicky, ite's kg ca aways caste kgside mauay with Ke1, Kf1 ad Kg1. But my feeg is that the kg is actuay safer where he sits, dead cetre. 23   as 24 bS! Botvik wsey deces to ope queeside es too soo. 24   cS?!  Let's assess the position:

: Why hnd Back a dbios ark when he ceary foows the prncipe: ix your opponent's pawns on the same colour



his remaining bish  ?

 This position is n exception. After this move Back's chnces dwnde nd vaish, snce he oses options of payng a kight to cS. The move effectivey throws Back's remang harmony nto a state of compete disarray.

25 b6! Itroducig a seaky itte expediet to isoate artificiay ad weake a. What a strategic squeeze . Now Back cat pay . . . Na6-c, sice he seaed c with a paw.

: Didn't Botv do exacty the

same by artificiay soatig his ow b-paw?

1. White's b-paw ist weak. 2. Back's f-paw remas a possibe probem shoud the game head for a edig. I fact, Back's defece of his advaced f-paw distracts from ay otio of pickig off b6. 3. By payig  paw to b6, Botvik separated the a-paw from hep, which is ow a secodary cocer for Back as we. 25  Ne8 26 ReI! Worryig Back with the possibiity of Re3, Bg4 (if ecessary), ad Rxf3. 26   Ng7 27 Re3 Qf8 28 Rb5!  He did, bt there are hge dfferences:





Back is competey tied up to defece of  weakesses o a ad f3. 28   Ra8 29 Na4 Fixg a ad opeg the potetia for Qc3, which adds aother attacker o a. 29  Qf7 30 Qc3 h5 31 Rxa5 Botvik's previous goodwi toward Back's weak paws comes to a abrupt ed. 31  Rb8 32 Nb2! Mutipurpose: 1. White may ted Nd3-el, addg yet aother attacker o f3. 2. White sowy cears the pathway for a3-a4-a, with the tetio of evetuay creatg a queeside passer with hs majority. 32  Kh7 33 Qb3

Pag Qa4 ad Ra8, with queeside fitratio. 33  Ng 34 Nd3 Nh6 35 Re : Why is Botv abandong his qest of f3?

 To ct the rose from the bsh, one mst be caref to navigate thorns. Botvk's refsa to take on f3, admittedy a perpexng decision, ndicates a concrete wa of obstnacy  bt keep n nd, takng the pawn too eary cod give Back unwnted counterpay on the f-fe. o Botvk, for now, ogicay ses the f3-pawn as a shied and concentrates his efforts on the qeenside.

35   Ng4 36 Qa4 Now Ra8 is  the air. 36  Qe7 37 Kc2 Rff 3 Ra7 N e 39 Bxg4 So f3 was ever wo. Botvik decides to eimate the kight ad keep the kgside as seaed as possibe. 39  hxg4

 (  ): Fnd a way for White to ake progress on the qeenside.

 Cear the way f or a3-a4-a5-a6, which creates a passed b-pawn.

40 Qb5! The pa migrates from the ream of thought ad imagatio to a reaity over the board. 40  Nf6 41 a4 Kg 42 Qa5!? There is o rush. Botvik marks time to ga time o the cock. Otherwise 42 a ooks more cosstet. 42   Qd 43 Nb2 Nd7 44 Ndl! Nf6 If Back takes o b6, ite seizes cotro after 44 ... Qxb6 45 Qxb6 Nxb6 46 Ne3, whe a rook is about to swg over to the ewy opeed b-fie ad g4 is doomed as we. 45 Qb5 Qe7 46 as Qh7! Eyeg e4, via a ... g6-g5 break. 47 Kd3! Coverg agast tricks agast e4. 47  Rf7 4 Qb2!? Patiece ca be take too far! It's as  Botvk grows tred of his domace ad

recosiders other optios. White shoud simpy coduct his guest up the board with 48 a6 bxa6 49 Qxa6. 48  NhS 49 Rgl Just i case Back  thikg about desperado sacs o g3 or f4. Botv k skifuy weaves ad eudes his pursuers, bockg off a aveues for potetia sacs. 49  g5!? Desperatio. Back hopes for a mrace by a quee fitratio to h2. So Smysov marches o, by ow obivious to the fiasco which ies ahead of him  his doomed edeavour. 50 hxg5 Rbf8 51 Qd2 Rf4!?





Now that's what I ca desperatio. Rare is the time whe perseverg eageress for a task effectivey suppats reaity. High feces make for suspicious eighbours, ad the message behid this move ca hardy be costrued as a peacefu gesture. Back's hoped-for revege is a ivg orgasm which must be fed to be kept aive. This fact, combed with his ife-threateg situatio, has a marveous way of shakg off apathy. I this case the predator grows bod  times of fame, whe prey  scarce, ad the rook presumes to approach White's kg as a equa. Botvik was udoubtedy  his habitua time pressure, so Smysov attempts to out-Ta his oppoet with oe of those hyper-charged momets, ike the spit secod before the other car impacts yours  a freeway coisio. 52 Nc3!? There was o accidet after a. Of course, Back's "sac exceeds creduity's boudaries ad Botvik coud take the rook as we: 52 gxf4 (White's g-paw, ike a reigio, toerates those who submit ad remas umercifu to heretics ad ubeievers) 52 . . . Nxf4+ 53 Kc2 Qxe4+ 54 Kb3 Qbl + 55 Qb2 Qd3+ 56 Nc3 ad Back's attack rus out of gas. But why aow Back ay fu at a? 52  Nxg3!? "I dot fear you! says the kight to the g3-paw, a itte too oudy. The kight fais  a drectios, ike a cat throw to a body of water. Admittedy pursug a rather famboyat ad ofty ambitio, the kight decides to go out  a baze of usoud gory, but othig ca save Back at this stage. Smysov fees obiged to fork over a exorbitat rasom to seek couterpay, which  the ed he is deie d.

53 Rxg3 Q2 54 Qel l-0 Smysov coud sti cotue his desperado ways with 54 ... Qxg3! (osg, but sti a good practica chace) 55 fxg3 f2 56 Qf1 Rf3+ 57 Kd2 Rxg3 58 a6! (58 Ne2 does the ob too) 58 . . . Rgl.

 ( ): Back is a hug amot of atria dow,

yet maages to geerate

serious threats o , ad aso to keep pushg with ... g4-g3-g2. hat woud you pay here for ite?

 Dfctio. Go dsprado right back at Back with th stg 59 RaS! ! (th rot 59 K2?? actuay oss to 59 ... xf1 60 Kxf1 g3 wh Whit has o dfc to th comg 61 ... g2+) 59 ... xf1 (or 59 ... xaS 60 Qxf2 ad Whit ws) 60 xfS+ KxfS 61 axb7 (ow Back is too sow  th rac) 61 ... Ra1 62 bSQ+ Kg7 63 Qc7+ Kg6 (63 . . . KgS is mt by 64 b7) 64 Qxd6+ Kxg5 65 Qx5+ ad Whit picks off th f2paw whichvr way Back's kg gos. Game 4 MBotvk-VSmyslov Word Champioship (2d atchgam), Moscow 195 King's Indian Defence  4 N 2 c4 g6 3 Nc3 Bg7 4 e4 6 5 f3

The Smisch Kg's Idia. S 000 0-0 6 Be3 a6! The preude to a remarkabe, ew (for the time) opeg idea. 6 ... e5 was the orma move for 1958.

7 Bd3 This sighty weakes hite's fuece over d4, so today 7 Qd2 ad 7 Nge2 are payed more ofte. 7 000 Nc6!!

·



Now it's Smysov's tur to rock Botvik  the opeg, with a (the) competey ukow Kg's Idia set-up which remas fashioabe to ths very day. The move is eve more potet whe White commits to a eary Bd3, sce he has eve ess cotro over d4 tha  orma es .

: What is Back's pa?

 Back pas both ... b7-b5 ad ... 7-5 at som stag, chagg ad chippg away at Whit's ctr simutaousy.

8 Nge2 Rb8 Back's true idea behd the e : he pas a queeside dsturbace vovg . . . b7b5, which may dscourage White from queeside castg. Istead, 8 ... e5 teds to veer back toward the ma es after 9 d5 Ne7 10 Qd2 c6 11 0-0 cxd5 12 cxd5, P.Haba-M.Vokac, Czech Champioship, Laze Bohdaec 1999.

9 a3 : Why pay a2-a3 - ?

 Botv wats to stop a futur ... b5-b5  its tracks with b2-b4.

9  Nd7 9 ... e5?! 10 d5 Ne7 11 b4 ooks good for White, sce Back's ... a7-a6 ad ... Rb8 make itte sese at this pot. 10 Bbl!

: This mov sms vry odd. What is Botvik's pot?

 Botv wats to kp firm cotro ovr d4. For xamp, i h payd th rot 10 0-0, it coud b mt by 10 ... 5 11 d5 Nd4! with a qua gam.

10  Na5

Targetig c4.

11 Ba2 b5 d oce agai. 12 cxb5 axb5 13 b4 Nc4 14 Bxc4 bxc4

: Which sid got th bttr of that ast skrmish?

 Nithr sid did, but th positio sharpd cosidraby. Back obtad th bishop par, wh Whit, who ow ows a passd a-paw, improvd his structur.

15 0-0 c6 16 Qd2 Nb6?! Smysov begs to drift. I dot see what the kight does o b6, other tha wait to get kicked by a3-a4-a. Istead, 16 . . . Re8! retas his bishop pair ad ooks more accurate. 17 Bh6! Botvik seizes hs chace . Principle Eliminate one of your opponent's bishops  he owns the bishop pair. 17 000 Bxh6!? Smysov is uafraid of a kigside attack ad fee ite's quee may actuay be offside o h6. 18 Qxh6 f6?! Ufortuatey, Smysov's maide voyage with his ew system was ot destied for a auspicious start. He waks away from his oce-promsig opeg ivestmet without profitig so much as a sou to his credit. I dot caim to uderstad this move, which weakes uecessariy. Back reazes, with a shock, that hs happiess, a mere abstractio, ever existed  the frst pace.

19 a4 Na Rug eve before he is hit  c7 is a ice square for the kight. 20 Rfbl f5?!

·



Soud advice s oy of vaue to those who heed it. Why s it that gaatry ad foy are aways so cosey itertwied? Ths move refects Smysov's emotios more tha the objective reaity of the positio. Smysov ooses hs game i a attempt to geerate activity ad sharpe the strugge. Soo the puget odour of hs decisio offeds  ostris, as a der of iver ad oios woud to a six-year-od who had couted o pizza stead. 21 Qe3?! 21 d5! ooks strog: 21 ... cxd5 (21 ... fxe4 22 Nxe4, threateg Ng5, is good for White as we) 22 exd5 e5 23 dxe6 Bxe6 24 a gives White a strategicay wo game, sice he domiates the dark squares ad soo estabshes a powerfu kight bockade o d4. 21 000 fxe4 22 fxe4 N c7 23 d5 Now Botvik gets a ess powerfu versio of the previous ote. Here 23 b5! cxb5 24 axb5 e6 25 d5 e5 (25 ... exd5?! 26 b6! Ne8 27 Nxd5 Nf6 28 Qd4 eaves Back struggg) 26 Ra7 was better, whe White begs to exert pressure o the queeside. He may be abe to pick up that stray c4-paw as we. 23 000 cxd5 24 exd5 Bb7 25 Rf1! Qd7 White's d-paw ist reay hagig: 25 . . . Nxd5?? 26 Qe6+ picks off a piece .

26 Qd4 Back's c-paw is i very rea dager ow. 26 000 e6 27 dxe6 N xe6 28 Qg4! Dreamig ad actig upo the dream are two etrey differet thgs, athough I do cocede that the former s a ot easier to achieve. Sti, a sge act surpasses a hudred dreams  such positios. The quee strikes the e6-kight a gacg, dsterested bow, just for show, the way a schoo buy picks o a smaer kid,  oy to advertise his ow buy credetias to the witessg chidre.

·



Botvik shows typica dsda for the materia ream, foregog Back's offer of hs c-paw  ad ow we observe a creased, quaitative shift  White's itiative. Less spirituay ced payers ike me woud grab a paw  a heartbeat with 28 Qxc4, which the computers say ist as powerfu as Botvk's ast move. 28  Rfe8 I this momet of hardship ad duress for Back, the good od days of froic ad comfort fee fugitive ad remote deed. Back cat hod the game after 28 ... d5 29 Nd4 Rbe8 30 Rxf8+ Nxf8 31 Qxd7 Nxd7 32 a5 Re3 33 Ncb5 sce hs passers are firmy bockaded, whe White's are free to roam forward.

29 Nd4 From a stadpot of carity there is somethg to be said for the simper e 29 Nf4 Nf8 30 Qxd7 Nxd7 31 Radl, forcg the w of Back's d-paw. 29  Qg7 30 Radl?! Now both payers foow with a series of secod-best moves, ad progress arrives  discouraggy sow motio. Mutua misfres ad faied uges beg to take a to o the cock ad erves of both sides. Much stroger was 30 Nxe6! Qxc3 31 Qf4!, whe White targets f7 ad d6. Back s busted here. 30  Nc7 After 30 . . . Qe5!, Back keeps his dsadvatage to a mimum. 31 Qf4! ReS? But after 31 ... d5 32 b5, White's paws ro oce aga, whie Back's go owhere.

·



It appears as if hts of modest reewa of prosperity ad piece activity beg to sprout  Back's camp, which s but iusio.

 ( ): mysov just bdrd  a arady

very difficut positio. How

did Botvik pouce o the error?

A Fork doub attack.

32 Nc6! The Fataffia kight,  a shockg breach of court etiquette, good aturedy saps both rooks across their backs, to the horror of the witessg roya . 32  Bxc6 33 Qxc4+ Doube attack. White regas the piece with terest. 33   d5 34 Qxc6 Rd8 Not 34 ... Rxb4?? 35 Nxd5 Rxd5 36 Rxd5 ad if Back recaptures with 36 ... Nxd5, the 37 Qe8+ mates ext move.

35 Qb6 After the commissio of a robbery, the experieced crima avoids rug, reasog that it is better to wak with a mmum of aacrity ad merge to the crowd without drawg udue attetio. Botvk chooses ths safer path ad bockades, rather tha rog his passers with 35 b5. 35  Qe7 36 Qd4 Estabshig the bockade . 36  Qd6 37 Rfel Rde8 38 Rxe5 Rxe5 39 b5 Fay, Botvik's passed paws move forward, furtive ad cautious as mice eterg a ufamiiar house. 39  Ne6 40 Qa7 d4 Ca you fee it? Other wordy forces foat  the wd, ookg to make troube for Back, who just broke White's bockade ad freed his d-paw. The oy probem: what Smysov beieves to be a race betwee passed paws is  reaity o race at a. We arrive  a situatio coducive to poitica trigues ad dark machatios.

 ():

How did Botvk da with Back's attmptd

dsruptio ad

idustriousy vacuum away further resistace?

A Th 4-squar is saf for th kight, who cotmptuousy swats th woud-b chag asid, as i dag with a aoyg sct.

41 Ne4! 1-0 ftr 41 ... QdS (41 ... 4?? fais misraby to 42 QaS+) 42 b6! (ow h ray dos sac th kight) 42 ... 4 43 b7, hit's qu tdry crads hr b-paw to swt drams o bS. Game 43 MBok-SFua Trag atch, Moscow 1961 King's Indian Attack 1

g3

Botvik s  a experimeta mood  this trag game agast Furma, who ater secured his pace  chess hstory by becomg Karpov's coach. 1 000 dS  1 ... e Botvik woud udoubtedy have payed 2 c4, eterg his favourite Egish variatios.

2 Nf3 g6 Of course,  such o-cofrotatioa opeg systems, Back ca set up how he peases. 3 Bg2 Bg7 4 0-0 eS S d3 Botvik gives Back the choice of reversed Kg's Idia or reversed Pirc. S 000 Ne7  . . . c eters reversed KID territory.

6 Nbd2 0-0 7 c4 7 e4 is payed more ofte here;  e. g. 7 ... c6 8 b4 a 9 bxa Qxa 10 Bb2 d4 11 Nb3 Qc7 12 c3 dxc3 13 Bxc3 c, whe chaces ook baaced, T.Radjabov-I.Cheparov, FIDE Grad Prix, Baku 2008.

7   d4 7 . . . c6 ad 7 .. . Nbc6 are ao possibe.

8 b4

8  as I thk it's sighty more accurate for Back to wait for a2-a4 before payg ... a7a.

: Why?

 It cuts dow o Whit's optios ad disaows a2-a3 atr o. For xamp: S . . . Nd7 9 a4 a! 10 b c 11 bxc6 Nxc6 12 Ba3 Nb4 13 Nb3 (o 13 Rb Back is happy to offr a paw for th bishop par d dark squars with 13 ... Nc! 14 Bxb4 axb4 15 xb4 Q7, wh I woud tak th back pics i giv a choic) 13 . . . RS 14 Qd2 BfS 15 fb Nc 16 Nxc Bxc 17 N Bd7 1S Nc2 Nxc2 19 Qxc2 Bxa3 20 xa3 Bc6 21 Bxc6 bxc6 22 Rab3 d Back's comg ... -4 cotrpay shoud offst Whit's cotro ovr th b-fi, E.Tomashvsky-Zhag Pgxiag, Russia vs. Cha atch, Nzhij Novgorod 2007.

9 b c 10 bxc6 : Dos't Whit's ast mov smpy assist Back's dvopmt?

 It dos, but Whit must op th qusid, his ara of cotrpay. If th qusid coss, th Back's ... -4, by dfaut, bcoms th oy viab paw brak  th positio.

10   N exc6! 11 Ba3 Nb4 12 N el! Destatio c2, to chaege the truder o b4. 12  Re8 Dua purpose : 1. Back dreams of a future vovg . . . e-e4. 2. Back aows for . . . Bf8 to support b4. 13 Nc2 Bf8 14 Bxb4 axb4 15 Nb3 Na6

16 e3!? A typicay bod Botvik decisio. He aows the game to ope, despite Back's bshop par, i order to ro his cetre forward. Prcipes dot aways impy restrictio. Sometimes we are compeed to faut prcipe ad vioate it to its face.

: What woud b a safr atrativ?

 Chip away at b4 by 16 a3 bxa3 17 Nxa3 with a qua but ss sharp positio, rsmbg a Bko Gambit structur  rvrs.

16 000 dxe3 17 fxe3 f5 is move weakes his kg's positio, which pays a roe  some future variatios. Perhaps Back's best e s 17 ... Nc5! (17 ... Bf5 18 Rxf5!? gxf5 19 Bxb7 offers ite exceet compesatio for the exchage) 18 Nxc5 Bxc5 19 Bd5 Be6 20 Bxe6 Rxe6 21 d4 exd4 22 exd4 Bf8 23 Qd3 with oy a ty edge to hite, due to hs extra cetra fuece .

18 Qd2 Rb8 19 a3

Removg a thor  his positio.

19  bxa3 20 Nxa3 Bb4 21 Qe2 Nc7 22 Nc2 Bc3 23 Ra3 b5 24 Nc5 Qe7 : Why ot ga a tmpo d support his bishop with 24 . . . b4 - ?

 Th advacd b-paw may grow wak aftr 25 Rb3, wh Na4! is  th air.

25 Rxc3 Qxc5 26 Qd2 bxc4 Hdini prfrs 26 ... Rd8 27 cxb5 Q7 28 Rc6 Nxb5 29 Nb4, ftr which Back ay ary b qua.

27 Rxc4 Qe7?! d here 27 .. . Qd6 s correct, whe White oy hods a miscue edge . 28 Qa5!





A cursory gace woud dicate equaity. However, upo deeper examatio we come to see ths is ot the case. Botvik uearths a hidde fudameta  the positio: Back's kight remas  dager. With this sge factor as a catayst, White's game bossoms with each passg move. The facie fashio  which he satches ad hags o to the itiative suggests a profoud mastery of the strategic essece of the positio. 28  Na6?! Back's siece o the matter takes o greater meag tha a agry outburst. Furma is uabe to varish over his difficuties this way. The kight, fdg himsef sged out for ridicue by his cassmates, sits aoe, head bowed ad arms caspg kees, whie Botvik cotues  attempts to corra it. Rather tha this stess poicy of actio, perhaps better was 28 . . . Nb! (the butterfy escapes the web ad ow his wgs oce aga futter ad kiss the sus rays) 29 Nb4 (threateg a fork o c6) 29 ... Qd6 30 Bc6 Bd7 31 Rfc1 Bxc6 32 Rxc6 Qd7 33 R6c, whe Back remas uder pressure after 33 ... Qa7! 34 Qxa7 Nxa7 35 Nd Red8 36 e4, but ist ost yet. 29 Rc6! The ster rook deigs to eucidate a importat esso to his impertet youg studet o a6. White adds a og to the fre of his pawess, queeside attack, uti it spits ad crackes to ife. Evets proceed with md-umbg swifess ad Furma is soo busted, eve before he appreciates his predicamet. Watch how Botvik cotuay seizes upo the a6-kight's discomfort. 29  Ra8 Back's kight, fraught with decsio, just sits there, refusg to tur eft or right. Furma must tread carefuy. If he attempts to free his kight with 29 ... Nb4??,

there foows 30 Rc7 Qd6 31 Nxb4 Rxb4 32 Bd5+ ! Kh8 33 Qa7 ad Back must give up a piece to cover h7.

30 Rbl Qd8 After 30 . . . Ra7 31 Qc3! Qd8 32 d4!, the white quees aurg dress fauts the cotours of her body, creatg distct dscomfort for Back's shy, togue-tied kg, who suddey ooks fatay exposed o g8. 31 Qc3! reateg Rbb6!. 31  Rb8? A mistake  though after 31 . . . Ra7 32 d4!, Back is uikey to save the game ayway.





Whe vague hope coides with the what-must-be, it is the former which must bow  acquiescece. Furma,  a aready ost positio, cracks uder the pressure. Now his sge, dreadfu apprehesio s give birth.



(

):

How

did

Botv

tak

advatag

of

th

doet,

ucommuicative kight, who cotues to brood o a6?

A Doub attack.

32 Rxb8! 1-0 Back drops a pic aftr 32 . . . Nxb8 (th kight, caught  a icit dd, hops to tak his way out  fortaty, his rvous stuttr ad rosy bush ark him as th most comptt iar aiv o th pat) 33 Qb3+! (doub attack) 33 ... Kg7 34 Qxb8. Game 44 MBok-MTal Word Champioship (15th atchgam), Moscow 1961 King's Indian Defence  4 N 2 c4 g6 3 Nc3 Bg7 4 e4 6 5 f3 0-0 6 Be3 c6 7 B3 e5

7 . . . a6 is aother pa for Back; e. g. 8 Nge2 b5 9 0-0 Nbd7 10 Rc1 e5 11 a3 exd4 12 Nxd4 Bb7 13 cxb5 cxb5, whe Back's active pieces make up for his sighty ferior structure, A.Karpov-G.Kasparov, Word Champioship (1st matchgame), New York/Lyos 1990.

8 Nge2 exd4 I thik a good opeg choice for Ta, who reveed i ope games ad was ceary Botvik's ferior if the positio bogged dow ito a maoeuvrig game.

9 Bxd4! Stroger tha 9 Nxd4 d5! 10 cxd5 cxd5 11 0-0 (11 e5? is met by 11 ... Re8 12 f4 Ng4 13 Bg1 f6, whe hite is  serious troube) 11 . . . dxe4 12 Nxe4 Nd5 13 Bf2 Nd7 ad Back ooked okay, J.Levitt-N.Davies, Presto 1989. 9  cS!? Ta agrees to weake d5  order to fight for d4.

: Why ca't Back pay 9 ... d5 hr?

 This tm th ... d6-d5 brak is't so hot for Back. ftr 10 cxd5 cxd5 11 5 Nfd7 12 f4 Nc6 13 Bf2 Nb6 14 0-0 f6 15 Bh4, Whit stads pasaty bttr, R. Hbr-.Gigoric, Lgrad Itrzoa 1973.

10 Bf2 Nc6 11 0-0 a6 12 Qd2 Be6 13 Radl QaS 14 b3 Rab8 15 B bl Rfd8 16 f4! Now f4-f5 is  the ar. Botvik seses a brewg kgside attack. 16  Bg4! Dua purpose. Ta sidesteps f4-f5 ad prepares to swap o e2, based o the prcipe: The cramped side should seek swaps. Istead, 16 ... Ng4? waks to the upeasat 17 Bh4, whie 16 ... b5? 17 f5 Bc8 18 Bh4! sees Back  deep troube.

17 3 Bxe2 18 Nxe2

·



Botvik switches gears ad aows a edig, reaizg that: 1. He ows the bishop pair. 2. Back must worry about a hoe o d. 3. Back must urse a backward d6-paw. 4. White ows the greater share of space. 5. Botvk most certay aso factored  that he was superior t o Ta  edgs ! 18  Qxd2 The back quee gives her sister a quick gace of cacuatio, mged with maice.

: If Ta was th suprior payr  compx iddgams,

the why ot keep quees o

the board with 18 . . . Qc7 - ?

 Ta corrcty swaps qus, raizg his activ pics ad cotro ovr th dark squars mak up somwhat for Whit's puss. Ta had a hathy rspct for Botvk's attackg skis. If h rtas qus o th board, h givs Whit autoatic kgsid attackg chacs aftr 19 f5! . Pas rtur to Chaptr O of this book, otvinnik on the Attack,  you do't biv m!

19 Rxd2 ReS Couterattackg White's weakest pot: e4. 20 Ng3 Bf8 21 ReI Re6 22 Nf1 Rbe8 23 Rde2 Bg7 24 g4!

Seg more space. 24  Nd7 Here 24 ... Nd4!? 25 Bxd4 cxd4 26 Ng3 Nd7 s perhaps worth cosiderg, sce Back's dark square cotro may fuy compesate for his weakeed structure .

25 Kg2 R6e7 Ta ca aso go for a more radica dark square cotro strategy with somethig ike 25 . . . Nd4 26 Rd2 f5 27 Ng3 fxg4! 28 hxg4 g5! 29 fxg5 Ne5 30 Bxd4 cxd4 31 Nf5 Nxg4 32 Kg3 Ne5 33 Rxd4 Nf7 34 Rd5 with Back compesatio for the paw, athough it was ever a wse idea to try ad hod a paw dow edg versus Botvik! 26 Nh2 Nf8! Itedg . . . Ne6-d4.

27 B4 Ne6 The back kights, restg vutures awaitg the death of prey, eye d4 with studious terest. 28 Rf1 Rd7 29 g5 h5! Looseg ite's grip over f6 ad temporariy hatg Ng4. 30 gxh6 Bxh6 31 Ng4! Bg7 32 Nf6+ Now Botvk gets two bshops for a pair of kights. 32   Bxf6 33 Bxf6

·



A bishop appears o Back's doorstep, as of yet a udefied, formess chaege, but a chaege oetheess.

: Is Back just osg hr? It ooks k Whit's bishops,

especiay hs dark-squared

moster o f6, wi be Back's udoig.

 ctuay, I thk Back hods his ow  this positio, ay du to his ow dark-squard grip o d4.

33  Ng7? The begg of a fase trai. The kight uwisey edorses a fatasy which he shoud have doe hs best to igore, despite the temptatio. Ta, very ratioay, fears the bishop's reig o f6 ad proceeds to remove it  but  dog so, he trasfers hs kight to a poor ocatio. Hs pa fees out of phase with the positio's requiremets.

: What pa do you suggst for Back?

 H shoud trust  his pic activity d pay 33 ... Nd4! 34 R R6 35 5 Kf8 36 h4 N7 37 h5 gxh5 (37 . . . Ng8 is mt by 38 Bh8!) 38 f5 (38 h Ng8 39 Bg5 f6 40 xf6 x 41 x Kf7 42 h Nxf6 43 Bxf6 Kxf6 44 Rxh5 Rg7+ is f for Back as w) 38 ... xf6 39 xf6 Ng8 with rasoab compsatio for th xchag.

34 Rd2! Aerty targetg d6 ow that Back acks a . . . Ned4 bock. 34   NhS 35 Bc3 Red

 (  ): How did Botvk strgth his positio?

A ctivat his oy dysfctioa pic, th ight-squard bishop, which h trasfrs to its optmum post o g4.

36 Bc2! Kf8 37 Bdl Ke7!? Uappetizig, as s 37 ... Ng7 38 Bf6 Re8 39 Bg4 Rc7 40 f5! gxf5 41 exf5. Back's positio is i disarray ad he soo drops a paw.

38 Bg4 The arrogat bishop, beievig himsef spirituay eevated above a others, decares to Back's started rook: "Bow before me, wretched itte ma! I am the ight! 38  Rc7 39 fS! Etombig the h5-kight. Aother briiat strategic decsio, correcty judgig that Back's cotro over e5 yieds him itte beefit. 39  Ke8 39 ... Ne5? is met by 40 Ba5, whe Back's irritated rooks remostrate with their ax guard, who ceary acks the ecessary dedicatio to protect his charges. 40 f6! A simpe pa, devoid of gitter, yet powerfu. ite's f-paw refuses to yied. Botvik's space advatage sps out of cotro ad Ta begs to asphyxiate. Oe begs to woder about the true et worth of the h5-kight, after its ackustre performace over the past few moves, ad who fds itse bereft of eve a sge viabe square. Now ite's g4-bishop cotues to gare dow o the competet uderg o h5 with discocertg tesity.

40  b5 41 Rd5 Gog after the straded h5-kight. 41 Rfd1! was eve stroger. 41   bxc4 42 bxc4 Rb7 42 . . . Nb4 s met by 43 Rg5! . 43 Kf3 Rb4! Ivetive pay ad a good practica try by Ta, whose tricks bubbe up, eve from the depths of hs wretched positio. The timid rook ches coser ad dares more tha he ever did before, refusg to se-veto his rash proposa. Geerosity ist oy the prerogative of the weathy. Sometimes the oy path to recocig a deterioratg positio with mutipe strategic woes is simpy to stop tryg to fix thigs ad go to war. I a positio brimmg with premoitory, oomg disaster, Ta attempts to seize fate by the coar ad steer it  aother drectio. He embarks o a desperate veture, but whe a ese fais, is it reay desperatio or just commo sese? I ay evet Botvk sees through the attempt to cofuse, which whips up froth but itte substace.

44 Bxb4 A coward has o quams about assaig a ormay more powerfu foe who has fae o hard times. 44  Nxb4





Hearteed by the tu of evets, Back maaged to offer sprited resstace. After

45 Rd2?! Nxf6, Ta cotiues to resist with a paw for the exchage ad dark square couterpay.

 (  ): Do you s a strogr pa for Whit hr?

A y sats mata pacfu quaty  tms of xtrm provocatio. Rturg th xchag yids a wo rook d paw dg.

45 Bxh5! As oted above, 45 Rd2?! Nxf6 46 Rfdl Ke7 47 e5 Ne8 aows Back to fight o. 45 000 Nxd5 46 exd5 gxh5 47 Rbl!





Back's reves come to a dsma cocusio, ad ow foows the hagover. This is the positio Botvik evisioed whe he retured the exchage. He targets a6 ad d6, Back's softest poits, whie h5 is o the critica st too. After precious ivestmet of toi, coi ad bood, Ta receives precious itte retur. His game, suffused i misery, ies beyod redemptio. Eve though materia is temporariy equa, there is o questio whether Back wi drop paws, ot with the massive activity dysfuctio of both kg ad rook versus hite's vigorous couterparts . 47 000 Kf8 48 Rb6 Kg8 49 Kf4 Givg h5 the evictio otice. 49 000 Kh7 50 Kg5 Rg8+ 50 . . . h4 51 Kf5! a5 52 Rb7 Kg8 53 Kg5 eaves Back hepess. 51 Kxh5 Rg3 52 h4 Re3 53 Rxd6 Re5+ 54 Kg4 Kg6 Or 54 . . . Re4+ 55 Kg5 Rxc4 56 Rd7 Kg8 57 h5 ad if 57 . . . Rcl, tedg to bombard the eemy kg with checks, 58 Kh6! pus a fast oe, forcg mate  two moves, as ite's tresome kg ad rook cotue to impose upo Back's kg. 55 Kf4 Rf5+ 56 Ke3 Rh5 57 Rxa6 Rxh4 58 Kd3

58  Kf5 Or 58 . . . Rh3+ 59 Ke4 Rh4+ 60 Ke5 Rxc4 61 d6 ad there  o race. 59 Rc6 Rh2 60 Rxc5 Rxa2 61 Rc7! Kxf6 62 Rd7 Ke5 63 Re7+ ! 1-0 If 63 . . . Kd6 (63 . . . Kf6 64 d6 is totay hopss as w) 64 xf7 Kc5 65 Rc7+ gts th job do. Game 45 TVPetrosa-MBotvk Word Champioship (1st atchgam), Moscow 1963 Nimzo-Indian Defence  4 N 2 c4 e6 3 Nc3 Bb4 4 Qc2 5 5 cx5 ex5 6 Bg5 h6

7 Bxf6 : Why did Ptrosia brak th tsio?

 H probaby watd to stay away from th Ragoz-k s stg from 7 Bh4 c, which Botvk probaby drstood bttr tha Ptrosia. Th 8 dxc5 g5 9 Bg3 N4 10 3 Qa5 is th usua cotuatio owadays, rathr tha 8 0-0-0 Bxc3 9 Qxc3 10 Bg3 cxd4 11 Qxd4 Nc6 12 Qa4 Bf5, wh Whit was arady  troub  Krs-Botv (Gam 25).

7  Qxf6 8 a3

Back must retur the bishop pair. 8 000 Bxc3+ 9 Qxc3 c6 10 e3 White hast exacty payed the positio ike a ba of fire ad it wot come as a surprse whe I assess it as equa. Back gas  freedom  though White sti chershes oe tiy hope: his queeside miority attack to-be. 10 000 0-0 11 Ne2!?  attempt to veer from a heaviy trodde path to a odd byway, which may be a bit too shifty for White's ow good.

: What is his ida?

Pa A) Ng3, f2-f3 ad e3-e4, as Botvik did versus Capa  Chapter Oe (Game 8). But here, too may pieces have bee swapped for White to have ay rea attackg chaces o the kgside . Pa B) Nf4, Be2, 0-0 ad a queeside mority attack, if give a chace.  Ptrosia wats to stay fxib ad prhaps atr pay for:

: What do you suggst stad?

 I prfr th straightforward 11 Nf3, gog for a qusid mority attack, as Karpov did  this gam: 11 . . . Bf5 12 B2 Nd7 13 0-0 fS 14 b4 a6 15 a4 (Whit ros his mority forward) 15 . . . R6 16 f, .Karpov-J.Pogar, Mot Caro (rapid) 1995. ctuay Back shoud b f hr, but Karpov maagd to w it ayway, just bcaus h is Karpov.

11 000 Re8! Beieve it or ot, this was a ew move at the time. Not a theoretica improvemets eed to be doube excams. Sometimes a simpe chage improves. I this case, Petrosia probaby expected 11 . . . Bf5, after which White itiates Pa B: 12 Nf4 Nd7 13 Be 2 Rfe8 14 0-0 Nf8 15 b4 Ne6 16 Nh Qg6 17 Ng3 Rac8 18 Racl Ng5 19 b5! with a edge  S.Reshevsky-E.Geer, Zrich Cadidates 1953. 12 Ng3 g6! A combative move  a o-combative positio. Eve du positios ca be payed eergeticay.

: What is th pot of Back's ast mov?

 Botv sists o a fight ad prpars th disruptiv ... h6-h5-h4.

13 f3?! As metioed earier, this pa does't quite fit the positio, sce ite is highy uikey to get away with a e3-e4 break  ad eve  he does, his attackg chaces wi be diuted by earier exchages. Botvk writes: "is move ca be uderstood oy if hite teded to caste og whe he woud frst have to guard the paw at bishop two (f2). I this case, Petrosia pas o such adveture, so his move simpy turs out to be a time-wastg weakess. ite shoud go for the simpe 13 Bd3 h5 14 Qc2 Nd7 (14 . . . Qxd4? !  met by 15 Bxg6! ad if 15 ... fxg6?? 16 Qxg6+ Kf8 17 Qxe8+! ws) 15 h3 Nf8 16 0-0 h4 17 Ne2 Ne6 with a eve positio, A.Beiavsky-Y.Baashov, Msk 1983. 13  h5! 14 Be2 I do't reay ike Botvik's suggested idea of 14 --?, after which 14 ... Qe7! ust embarrasses hite. For exampe, 15 Re h4 or 15 Rd3 h4 16 Ne2 Bf5 picks off the e-paw, whie 15 Kd2 hags o to the paw but is just askg for it! 14  Nd7 15 Kf2 h4 16 Nfl





ite's game  a portrait of ethargy. Back ow hods a edge . 16  Nf8 17 Nd2 I order to deveop  h-rook. 17  Re7 Preparg to doube rooks to crease the pressure o e3. 18 Rhel Bf5 19 h3!? Probaby payed to reease himse from the worry of ... h4-h3 at every stace. O the dowside, the move creates a weakess aroud ite's kg. 19   Rae8 20 Nfl Ne6 21 Qd2

·



Petrosia refuses to itiate a miority attack, correcty sesig that ay opeg of the game woud favour Back. So he sticks to his pa  do othig at a! I this positio Botvk cosidered but the rejected the pa of 21 . . . Ng5, aimig to sac a bshop o h3. So we arrive at a deep divide  how to make a ivg? Hoest mercatism (21 . . . Ng7) or crime (21 . . . Ng5)? Oe caot auch a attack vicariousy,  thought aoe. Does the idea work? Or s it better to ay such twisted ambitio aside?

work out the ramificatios. If you beieve the sac works, the back it up with cocrete es. Warg: This is oe of the most difficut exercises  the book!  ( ): Tak 15 to 30 uts ad try your bst to

21 000 Ng7?! A time pressure misfre more tha a error, but aso a gigatic ost opportuity. The deep secret of the combatio trembes, tataizgy ear yet far away, withi Botvik's cosciousess. Now Back's efforts ead to a accumuatio of oddmets more tha actua tagibe advatages, ad irresoute dreams, rather tha a actua pa. Botvk prepares a trasfer to f5 or h5, which retas his advatage.  But much strogr was 21 ... Ng5!! (th sociopath o g5, dvoid of ora hum motios, mry pays at f, at ormacy  it works!  th thrat is .. . Bxh3! ) 22 Kg (ot 22 Rac1?? Bxh3! ad f 23 gxh3?? N4+! ws th qu) 22 ... Bxh3! (ayway!) 23 gxh3 Nxh3+ ad ow:

a) 24 Kh2 Rxe3 ! ! 25 Nxe3 (the kig kight ht eyes the truder truder with outraged detes det estatio tatio ad orders him seized; 25 Kxh3?? gets crushed after 25 ... Rxe2!) 25 ... Qf4+ 26 Kh Nf2+ 27 Kg Kg Qg3+ Qg 3+ 28 Kfl Nh3 ! 29 Bdl B dl Qg + 30 Ke2 (the kg's kg's stoma s tomach ch kots  dark premoitio) 30 ... Nf4 mate. b) 24 Kh Qg5 ! 25 Kh2 Qg + 26 Kxh3 (the aged, firm firm kg edeavours edeavou rs to wak o hs ow power, eve  oy oy a few few steps) steps) 26 . . . Rxe3! Rxe3! .

Hdini assssmt: -9.39! Botv wrot: /t was dfficut to prciv that, dspit his two pics up, Whit is hpss ad dos't hav a sg satisfactory mov.

hat bows my md is that Botvik assimiated the attackg requremets ad the coated them sequet se quetia iay y to perfectio pe rfectio  his aayss aayss exacty e xacty matched matched Houdini'  but hs hs misasse misassessme ssmett of the the variatio variatio  threw threw him off! Chess Chess proves to be a miseraby tough game if a payer ca see a this ad sti go wrog!

: I sti do't s it. How dos Back w f Whit dos othg?

 Lk this: 27 Rac1 R84! 28 Rcd Rg4! with at  two movs.

22 Radl NhS 23 Rcl R cl Qd6 Q d6 24 Rc3 Ng3 25 KgI NhS Just testg! I such positios there is o rush. Botvik tacks about, fshg for a attackg attackg idea. idea . 26 Bdl Re6 27 Qf2 Qe7 28 Bb3

Cotempati Cotempatig g e3-e4 e3 -e4.. 28  g5!





At ast ast  motio. motio. Botv Bo tv ik ik ited tedss . . . Bg6, Bg6, foowed by . . . f7-f5 f7-f5 ad ad the the either either . . . f5-f4 or ... g5-g4. At ths poit 28 ... Nf4, with dua threats o d3 ad h3, is simpy met by 29 Qd2.

29 Bdl Bdl Perhaps the time has come for White to take actio himse with 29 e4 Bg6 30 Ne3 Nf4 31 exd5 Nxd5 32 Bxd5 cxd5 33 Qd2 f5! 34 Kf2 Qd6 35 Nc2 f4 36 Rxe6 Qxe6, which which ooks about eve. eve . 29  Bg6 Bg6 30 g4? g4??? Straightforward, bod, direct ad competey icorrect! Petrosia has had eough ad decides to right what he beieves to be a batat wrog, abadog himse to the puge. He refuses to exted trust  his waitg tactics,  what he beieves to be a faig eterprse for himsef o the kgside. But it's impossibe to overstate the overty braze (ad correct!) ature of this move. The positio, for so og weighted with with im imm me ece, ce, ow erupts  o Whi White te aoe aoe ! Petrosia Petr osia,, cast cas tg g a apprehesive assessmet, decides vigorous coutermeasures are a ecessity ad gambes correct correcty. y. Botvik wrote that he was  time troube here. Was this a paraoid reactio o Petrosia's part, or did he get greedy ad succumb to a opportuistic urge to muck the game up? The troube s that White faied to compicate. Istead, he merey weakeed weakee d his ow positio pos itio suiciday.

30   hxg3 hxg3 31 31 Nxg3 Nxg3 Nf4! Nf4! After Petrosia's batat vioatios, he must ow dea with harsh exactios imposed by the aw. "This move was apparety overooked or uderestimated by White. But I ask quite reasoaby: how is it possibe that a payer of Petrosias titaic payg stregth actuay overooked such a simpe move (which most A-eve payers woud see) with pety of time o his cock? The aswer  that word champioship matches, with their coceivabe ervous tesios voved, have their ow quat rues of who buders ad whe.

3 2 Q2 32 Qf1  met by the crushg overoad trick 32 ... c!.

 ( ): Back has two paths to th w



oe simpe, the other compex.

Fd oe (or both) of them.

A ovroad. ovroad. Ev  tim prssur Botv pays th forcfu  ovr th smp o! uch ar th ways of a prfctioist!

32  cS!

 #: Th simp trfrc 32 ... Nd3! crushs, sc hit drops th 3-paw.

33 Q d2 Petrosia sees that other moves ose without a fight: 33 dxc5 d4! or 33 Rxc5 Rxe3. 33  c4!? 33 . . . Qf6 Qf6 ! ws ws o the the spot after 34 Rxc5 Rxc5 Nd3 or 34 dxc5 dxc5 d4. 34 Ba4 b5!





Eve i dreadfu time troube, Botvik with the itiative was a moster.

35 Bc2 Or 35 Bxb5 Rb8 36 Ba4 Nd3 37 Rbl Rxe3 ad ite coapses. 35    Nxh3+ Nxh3+ 36 Kfl Qf6 Qf6 37 Kg2 Kg2 Nf4+ Nf4+ 38 exf4 exf4 The equivaet of resigatio, though ateratives are of o service either; e. either;  e. g  g. 38 Kgl Bxc2 39 Rxc2 Nd3 40 Rc3 Qxf3 or 38 Kf2 (the cofused kg staggers about, uabe to differetiate up from dow) 38 ... g4!  crushig. 38  Rxel 39 fxg5 Qe6 40 f4

 ( ): Thr ar a mio ways to w hr.

Fd the path which simpifies to the easiest victory.

 impficatio.

40 000 Re2+! 0-1 Th rook codscds to gc at Whit's qu ad kg, both chi d arrogat. 41 Nx2 Qx2+ 42 Qx2 x2+ 43 Kg3 xc2 ws triviay. How wodrfu wh o's wath is augmtd with v mor wath. Game Game 4 k-GS Bo  Bo k-G S zlagy IBM Touramt, mstrdam 1966 King's Indian Attack  g3 5 2 Nf3 c6 3 Bg2 Bg4

This  how I ormay pay the back pieces agast hite's set-up. The reversed Torre is oe of Back's most reiabe set-ups.

4 d3 Botvik opts for the Kg's Idia Attack. 4 000 N d7 5 h3 Bxf3 : Why had ovr th bishop par without a fight?

 ftr 5 ... Bh5 6 g4 Bg6 7 Nh4 Whit gts th bishop par yway. Howvr, Back did maag to duc h2-h3 ad g3-g4, which coud atr prov a wakss for Whit. o  ssc, I agr that Back shoud hav backd his bishop to h5 rathr tha chop o f3 mmdiaty.

6 Bxf3

6 000 e5 Ad Ad here I woud wo ud hod back the e-paw to e 6, formg formg a c6 c 6 d5 e6 wa desig de siged ed to hider hider hite's ig ightht-squ squared ared bishop. bishop .

: You woud't tak th ctr f offrd?

 Back's ast mov is tmptg but subty waks th ight squars.

: What ight squar wakss?

 tay td ad watch how Botvk gos about dog just that.

Istea Istead, d, 6 . . . Ngf6 7 Bg2 e6 8 0-0 Bd6 9 e4 Qc7, as   A. Morozevic Moro zevichh-V V . Krai Kraik, k, Mote Caro (bdfod rapid) 2007, is how I woud pay the positio as Back, who has rid himse of hs bad bishop ad achieved decet midde game chaces.

7 Nd2 N6 8 e4 dxe4 Back ca ao reta the paw tesio with 8 ... Bc, but the he must be o the costat ook out for e4xd, which opes the positio for White's bishop par. 9 dxe4 dxe4 BcS BcS l0 0-0 0-0 Qe7 Qe7 11 c3 c3 Thikg hikg about que q ueeside eside expasio exp asio with b2-b4. 11 000 0-0 It may have bee prudet to toss  11 ... a, as  Hartoch-D.Marovic, Wijk aa Zee 1968, whe Back curbs White's ambitios of expasio o the queeside for ow.

12 b4 Bb6 13 a4 Rfd8 Here I woud chae chaege ge Whi White te oce aga aga with with 13 . . . a 14 Ba3 Qe6 .

14 Qc2 Rac8 This move by itsef ist so terribe, but it is the beg g of a corre correct ct pa. 15 Be2! Botvik seses Back's comg erroeous pa ad trasfers his bishop to cover c4. 15 000 cS?





The positios reaity is ceary at odds with Sziagyi's msperceptios of what costitutes truth. Back evsages a happy future of abudat couterpay, peace,

prosperity pro sperity ad ogevity ogevity  oe of whi which ch come to pass. pas s. T makeshift makeshift attempt at a queeside / attack resembes actors payg at war, rather tha war itsef. 15 ... c s a serious strategic error, which by today's stadards woud easiy deserve a /77"

: Why? It ooks thatic. Back fay chags Whit o th qusid.

 Th mov vioats th prcip: Don't x your pawns pawns on the same colour This gam is dstd to bcom th postr chd for this prcip's vioatio!



your remaining bish.

16 bS Ne8?! Back wats to pay ... Nd6 ad ... c5-c4 but Botvk wot give him the time. Beieve it or ot, Back shoud be  desperatio mode aready ad simpy give  c paw away with 16 ... c4! to cear c ad orgaize pay dow the c-fie. I ths case he s sti probaby busted, but at east he does't suffer the joyess, couterpayess, dystopia future which which esues    the the game. 17 Nc4 Nc4 Nd6 18 BgS!





 aert shot which further weakes Back o the ight squares. square s.

18  f6 f6 Udersc Uderscor org g Back's Back's dwd dwdg g f fuece uece over the ig ight ht squares. square s. 18 . . . Qxg5 19 19 Nxd6 Rb8 20 Bc4 Rf8 Rf8 21 2 1 Radl may be eve e ve worse, whe whe Whi White te domates domates..

19 Be3 Nxc4 : How about about 19 . . . Nf7 - ? My ida: Whit Whit ows ows c4, c4, so by

eavg the kight o that

square, Back deies the bishop use of the al-g8 diagoa. I essece, White has two powerfu pieces whic which h wat wat to occupy occ upy c4, but he ca oy pay oe to the square s quare..

 Viab pas for Back  this positio ar fugitiv. Lack of cotrpay vsts his positio with a spok, drab strity at th mptss of futr prospcts. Yor cocpt is cvr but Whit's advatag grows  ay cas aftr 20 Qb3 Bc7 21 a Bb8 22 Bg4 g6 23 h4 Kg7 24 fd1, wh Back is  a bad way. I th it's too at for Back. H sad his fat stratgicay with his horrib 15th mov.

20 Bxc4+ Kh8 21 as Bc7 22 Rfdl Nf8 23 Qa2!

·



A disarmigy simpe, yet powerfu strategic stregtheg. White creases his grip o the ight squares, pre vets . . . Ne6, ad covers ad reforces a which frees the a-rook for actio. Back's game is beyod redemptio. We take vetory: 1. White ows the bshop pair. 2. White domates the ight squares. 3. White hods a grip o d5 ad c4. 4. Back ows a worthess bishop, hemmed  by hs ow paws (ot to metio a rather ame kight). The bshop may be a exceet repica of a rea, fuctiog piece, but o repica matches the prste perfectio of the origa. 5. White ows a gigatic territoria advatage o the queeside. Cocusio: Back is strategicay ost. 23  Rxdl+ Hopg that swaps wi ease  defesive troubes. I ths case, they merey hep Botvik, sce Back weakes his back rak. 24 Rxdl Rd8 25 Rxd8 Bxd8 26 a6 b6 No choice, sce capturg o a6 is eve worse.

27 Kg2 Qd7 A hawk hauts our street. She (or he, I oy assiged the geder by tuitive defaut) sits atop a od teephoe poe, her rovg eye surveyg her doma for potetia prey. Oe time she swooped dow o a ufortuate pigeo, who was merey mdg his ow busess  our frot yard, ad scooped it up to her air atop the poe. Soo it bega rag feathers to our yard (which freaked out my wife big time!). We,  this positio Botvik pays the roe of the hawk. No oe questios that White domates the ight squares which, up uti ow, have bee pretty to ook at, yet produced itte tagibe ga. To stea is to obta that which is ot yours through surreptitious meas; to cofiscate is to stea opey, from a positio of power. I this stace, we see a exampe of the atter, where Botvik cofscates the ight squares across the board (aog with Back's hopes), assumg etera stewardship.

 (  ): Com up with o dvastatg ida ad

Back's word is abaze i the fire

of the ight squares.

A Qu ad bishop smpy trad pacs, aftr which Whit thrats Back's back rak, as w as mutip fitratios. This was th dowsid to Back's swappg rooks. But i h did't do it, th Whit woud hav itiatd th rook xchag hmsf.

28 Qe2! The quee, cotrivig a pa, thks to hersef: "That which I faied to achieve through war, I wi ow acquire through craft. 28  Ng6 29 Bb3! Ne7 30 Qc4! The creature, so og dormat o a2, awakes to her ow setiece ad power. A specuatios are resoved: White pays directy for mate. 30   h6 31 Qf7 The quee strides forward, with oy her eyes offerg a timatio of immet vioece. 31  Kh7 The kg cautiousy pe ers forth, a ervous rabbit from his hoe  a forest rife with predatio.

32 Bc4 Qd6 The streets o the kgside are empty of habitats, other tha Back's kg, ad the oy soud heard is the rustg of eaves  the wd. I the gare of suight it's easy to miss the cracks  a evromet of texture ad shadow.

 (  ): W arriv at Part 11 of Botv's ida: wav

matg ets based o the

ight square domatio. Try ad come up with two formatios which ead to Back's eary demise.

A Matg ts:

1) h3-h4, f2-f3, g3-g4 ad the Bxh6!. 2) h3-h4-h5, Qe8 (if aowed), Bf7 ad Bg6+. 33 h4! Qdl If Back shuffes ad waits, hs ed arrives with 33 . . . Qd7 34 f3 ! (matg et #1) 34 . . . Bc7 35 g4 Bd6 (35 . . . Kh8 36 g5 fxg5 37 hxg5 s ao curtas for Back) 36 Bxh6! Kxh6 (the kg sumps  despodecy, ike a partiay fied sack of gra, fodg forward) 37 Qh5 mate. 34 Qe8! Matg et #2 it s! The quee dazzes with her feme wies: tits of the head cacuated to etice, sham pouts, actressy dispays of mged gee ad chagr, ad tataizg, mock auremets too umerous to metio. 34  £5 Or 34 ... Qd6 35 h5! Bc7 (the shuffg adroid bshop, to  horror, reazes hs/its true ature: devoid of spirit, md or idetity  merey a copy of some thg ese) 36 Bf7! ad Bg6+ is comg.

35 exf5 Nxf5

·



Otherwordy, tremuous forms appear from the mst, foatig  the drectio of Back's kg.

 ( ): You gt tossd a

easy oe! hite to pay ad force mate.

 No xcam rqud for this! 36 Bg8+ Kh8 1-0 c 37 Bf7+ 7 38 Qg8 is mat. Hav you vr trd a mpty, abadod hos, yt ssd th occupacy of ow-dad prvious owrs? Game 47 Bo k-B Lae Paa d Maorca 1967 Rti Opening 1 c4 Nf6 2 Nf3 e6 3 g3 5 4 Bg2 Be7 5 0-0 0-0 6 b3 c5 7 Bb2

I ted to pay 7 e3 at ths pot.

: Why woud you mak this mov ovr

the fiachetto of the bishop, which

deveops a piece? 7  Nc6   prsoa prfrc. I do't wat Back to toss  7 ... d4, btg th b2-bishop. Whit shoud cotu 8 b4 a 9 bxc Nc6 10 d3 , .Mis-E.Gr, Wijk aa Z 1977, wh th positio is prhaps car aftr 11 Na3. Whit's op b-fi ad b-outpost compsat for Back's xtra ctra cotro.

8 e3

·



Now the positio  back to orma aga. 8  b6 : Why ca't Back pay 8 ... d4, rturg to th positio you fard?

 It's ot quit th sam. For xamp, aftr 9 xd4 cxd4 10 R N8 (tdg . . . f7-f6 ad . . . 6) 11 Ba3 f6 12 Bx7 Qx7 13 a3 a 14 d3 , w rach a odd, rvrsd Boi, E .Tomashvsky-M.Narciso Duba, Europ Champioship, Povdiv 2008.

Aterativey, 8 ... dxc4!? aso creates a immediate imbaace. D.Howe L.Paichadze, Europea Champioship, Povdiv 2012, cotued 9 bxc4 Qc7 10 Nc3 a6 11 Rc1 Na5 12 Qe2 Bd7 (Back ormay pays for ... b7-b5) 13 e4!? (a terestg terpretatio of the positio by Howe. 13 . . . Rfd8 14 e5 Ne8 15 Ne4 b5 16 d3 Bc6 17 Bc3 Rd7 18 Rfd1 Rad8 19 Rd2 b4 20 Bb2 Nb7 21 h4, whe I ike ite's chaces o the kgside.

9 Nc3 Bb7 10 d3 Accordg to curret theory, 10 cxd5 is ite's best shot at a edge, whe Back may respod  two ways: 10 ... Nxd5 11 Nxd5 Qxd5 12 d4 Qd8 13 Ne5 Nxe5 14 Bxb7 Rb8 15 Bg2 ad ite's bishop par gave him the edge, V.Kramk-T. Radjabov, Cadidates match (2d bitz game), Kaza 201 1. Or the sharper 10 . . . exd5 11 d4 Ne4 12 dxc5 Nxc3 13 Bxc3 bxc5 14 Qe2 Re8 15 Rfd1 with a sharp strugge ahead agast Back's hagg paws, V. Kramik-G.Joes, Lodo 2012. 10  Rc8 11 Rcl Rc7 I woud go for 11 ... dxc4 12 bxc4 Nb4.

12 Qe2 Rd7 I sti beieve Back gets dyamic equaity after 12 ... dxc4.

13 Rfd1 ReS

: Is't Lars payg ovry cautiousy this gam?

 It crtay appars that way. H putzs arod for too og ad ow Whit gts a dg. Istad, Back ca try th mor dyaic 13 ... d4 14 xd4 cxd4 15 Nb Bc5. But do't b food to thg that Lars was mry payg for a draw. By 1967, h was strogr tha Botv, who was w past his prim ad at th ta d of his carr, wh Lars was just trg his hyday as a adg chagr for th word champioship.

14 cxd5 Nxd5 14 ... exd5!? drops a exchage to 15 Bh3, whe I thik Back fai to get fu compesatio, despite hs soo-to-be fearfu ight-squared bishop.

15 Nxd5 Rxd5 Larse was rightfuy fearfu of eterig ay kd of hagg paws situatio agast Botvik, arguaby the greatest practitioer of that structure of a time. 15 . . . exd5 16 d4 woud ead to positios Botvik uderstood better tha ayoe ese .

16 d4 Back's rook is a target o d5. 16   Qa8 17 dxc5! Pnciple: Open the position when leading in deelopment



v wh oy sighty adg, as  this cas.

17   Rxdl+ 18 Rxdl Bxc5? Natura moves are ot aways the best. Larse uderestimates the dager to his kg. 18 ... bxc5 was ecessary. 19 Ng5! reateg Qh5. Principle Create conontation when your opponent's fors are oside and out of sync Suddey, Back's pieces ook far away o the queeside, ad the approachg kight, by defaut, s viewed by Back's kg as a source of utmost dstrust. 19  h6

 ( ): ftr Back's ast, forcd mov, Whit has a

asty shot. Let's see if

you ca fd the move two chess egeds missed.

20 Ne4  Obitrat th kg's covr . Th txt is sti strog, but Botv had 20 Nxf7! ! Kxf7 21 Qg4! (v mor accurat tha th chck o h5) 21 . . . Nd4 (ot 21 . . . BfS? 22 B4! N7 23 Qh5+ KgS 24 Bxb7 d Whit rgas th pic with a crushg positio, wh i Back tris to hag o to vrythg, h gts matd ftr 23 ... g6 24 Q5! Bx4 25 Qf6+ KgS 26 Qx6+ 7 27 Qf7+ tc) 22 xd4 BfS 23 Bxb7 Qxb7 24 R with a ca xtra paw d tiativ for Whit.

20  Bf

 (  ): This is

possiby the most difficut

exercse  the etire book! Take at east 15 mutes ad cacuate 21 Nf6+ as deepy as you ca. Does it work?

21 Rd7?!

 It works! Botvik missd what woud hav b o of th gratst combatios of his iustrious carr: 21 Nf6+!! (wh avs bg to shivr  th forst o a wdss day, it ca oy ma o thg: a prdator approachs) 21 . . . gxf6 (othrwis Whit smpy picks off  xchag) 22 Qg4+ Kh7 23 B4+! (a avot prsc ficts itsf upo Back's kg, as bams of ight brush up ad dow his body, ad h raizs: th mothrship scas m") 23 . . . f5 24 Rd7! (a v quickr mat tha 24 Bxf5+! which forcs mat as w) 24 . . . N7 (or 24 . . . R7 25 Bxf5+ with siiar pay) 25 x7! (th mob boss may ordr th hit o Back's kg, yt it is his capo ad crw who gt thr hads dirty by prforg th actua dd) 25 ... x7 26 Bxf5+ (somtims forcd movs ar good os too) 26 ... xf5 27 Qxf5+ Kg8 28 Qf6! Kh7 29 g4!!.





Sarig Back's kig, whose fragie skeeta system was ot desiged to bear the weight of the rubbe crashig dow aroud his ow broke body. This move is impossiby difficut to see whe begg the combatio! Back has o defece agast 30 Qh8+ Kg6 31 Qg8+ Bg7 32 Qxg7 mate. The masterpiece combatio which ever was! Had Botvik foud this over the board, this game woud be promety posted  Chapter Oe: Botvinnik on the Attack! I remember oce I payed o a juge gym  the park as a kid at the very highest eve. I ost my grip ad fe  stages, boucg off each eve, uti I hit the grass beow with a sickeg "thud soud. Simiary,  ths variatio, Larses pa arrives  stages. I guided the positio from move 21 to my 2400 rated, 15-year-od studet Varu Krisa ad, without movg the pieces, he foud 29 g4! !, provg that ths is ot just some esoteric, theoretica exercise for the computers. If you practice these very difficut meta chaeges, you get better ad better over time. 21  £5? 21 . . . e5 was absoutey essetia to Back's surviva. 22 Nd6! Eve stroger tha 22 Qh5!, which aso ws after 22 . . . Re7 23 Nf6+ ! gxf6 24 Qg6+ Bg7 25 Bxc6! Rxd7 26 Bxd7 Qd8 27 Bxe6+ Kh8 28 Bd4. 22  Bxd6 If 22 . . . Re7?? 23 Rxe7 Bxe7 24 Nxb7 Qxb7 25 Qc4 ws a piece .

·



This is my tweth chess book for Everyma ad I dot remember a sige game where I cuded two critica decisio exercses  a sge game. Ths game proves to be the exceptio.

 ( ): BotvU is facd with a agozg

choice. Take o g7,  effect

sac'g a piece, or recapture the bshop o d6, with tremedous strategic cotro. Both paths ead to advatage, but oe is fitey stroger. What does your tuitio te you?

23 Rxd6?! Od age makes foos of us a. We see that, at ths very ate stage  hs career, Botvik's oce egedary powers of cacuatio have degraded. Oce aga, he faters whe o the cusp of a kockout bow.  Whit's bishop ad rook suprimpos thir doac upo g7. Hr 23 xg7+! givs Whit a dcisiv attack aftr 23 ...  24 h7! !, thratg Qh5. This ast mov is vry dfficut to spot. Oddy ough, Back is paraysd ad ca't mprov his positio a iota dspit th fr mov. Hdini assssmt is +16.43! Back ca rsig.

23  Nd4! To the crima md, the cocept of hoesty s but a abstractio. Larse cevery cotrives a method of compicatg his way out of his predicamet, or so he hopes.

24 Rxd4 Bxg2 25 Rd7

25  Bh3? The master of misdirectio oy maages to cofuse himsef. The bshop pursues a dstorted refectio of the positios reaity. Here 25 ... Bh1! puts up far sterer resistace, foowig the pricipe: Opposite-coloured bishops favour the attacker After 26 f3! (26 f4 is met by 26 ... e5! 27 Bxe5 Rc8) 26 ... Qxf3 27 Rxg7+ 8 28 Qd2! (threateg a huge check o d6) 28 ... e5 (ot 28 ... Qd5? 27 Qb4+) 29 Rxa7 Qc6 30 Rd7 Kg8 31 h3!, ite shoud sti wi with exact pay, but it woud be far more difficut.

26 £3 Now Back's bshop remas impouded to the warehouse o h3, to sit i the dark ad coect dust. 26   R d8 27 Rxg7+ Kf8 28 Rh7 Qd5 29 Kf2 Qd1

30 Rh8+ The huma move, eadg to simpificatio ad a tecicay w g edg. Houdini sists o 30 Bc3! which, see through dispassioate computer eyes, is deed stroger. 30  Kf7 31 Rxd8 Qxd8 32 Qc2 Qd5 Perhaps thikg about . . . Qb5. 33 Qc7+ Ke8 Certay ot 33 . . . Kg6?? 34 Qg7+ Kh5 35 Qf7+ Kg5 (the agie kg thiks to

himsef: "I am ot about to be take hostage without a vigorous chase, which iroicay turs out to be the fia thought of hs ife) 36 Bf6 mate. 34 Qb8+? Kd7? Mutua time pressure errors. Back's fortitude coapses ad, with it, a resistace as we. Istead, 34 ... Qd8! (moarchs merge aogside each other ike docked cruise ships at port) forces quees off the board ad shoud hod the draw. 35 Qxa7+ Kc8 36 Qa6+! ite's quee gets back to defed agast the . . . Qd2+ threat. 36   Kc7 37 Qc4+!





Botvik fees a marked sese of reief, the same way I did o December 22d, 2012  the day after the oe the Maya caedar predicted the word woud ed. It took icredibe judgmet to see that hite does deed w here, eve with two extra paws. Houdini coudt sove it ad gives ite oy a sight pus, whe  fact Botvik aready worked out the w. 37  Qxc4 38 bxc4 Kc6 39 Bd4 Oh, o you dot! Botvk deies Back's kg etry to c. 39  h5 40 a4 Zugzwag! Back's kg must give way, aowg hite his first passed paw. 40   Kc7 41 c! bxc5 42 Bxc5 Kc6 43 Bb4 Kb6

 (   ): Back's kg ad bishop

stridety protest agast the

umerous idigities to which they have bee subjected. How does White make progress?

 Forc kg try o th kgsid by sac'g a paw.

44 g4!! Botvik's move s far more cear tha the rote 4 4 a5+ Kb5 45 Bc3 Ka6 46 Ke2 Kb5 47 Kd3 Bg2 48 f4, whe Back gets drawig chaces, despite  two paw deficit. 44  hxg4 44 . . . fxg4 45 Kg3 (threateg f4-f4, e3-e4 ad f4-f5, creatig a kgside passer) 45 ... e5 46 Bc3 does the trick. 45 Kg3! Back's bishop becomes o more tha a specime seaed  a gass case, to be spected ad experimeted o at White's wi. Both back kg ad bishop fd themseves squeezed, uabe to move forward or back  much ike Ha Soo, Luke Skywaker, Prcess Leia ad Chewbacca, whe they were trapped  the trash compactor aboard the Death Star. 45  e5?!

: This ooks ik suicid, pacg a paw o th sam coour as

White's remag

bshop. How does he make progress  Back does othig?

 Lars's ast mov maks it asir for Botv, but h st ws i Back simpy waits: 45 ... Ka6 46 fxg4 fxg4 47 B K6 48 4 Ka6 49 Kf4 K6 50 a5+ Ka6 51 5 K5 52 Kg5 Ka6 53 Kf6 g3 (or 53 ... K5 54 Bg3! Kxa5 55 Kx6, wh th -paw costs Back a pic, ad v mor mportaty, Whit's bishop is of th corrct coour for his rag h-paw) 54 hxg3 Kb5 55 Kg5 Bg2 56 g4 Bf3 57 Kf4 Bd 58 g5 Bc2 59 Bd2 Bh7 60 K3 Bg6 61 Kd4 Bh5 62 a6! (dfctio) 62 ... Kxa6 63 Kc5 Bg6 64 Kd6 Bf5 65 K7 ad th g paw costs Back his bishop.

46 e4!





Botvik keads the soft structure to the shape of hs desires. 46  fxe4 46 . . . f4+ 47 Kh4! (the kg comes to a decsio, thkg: "if the rabbe refuse to ove me, the et them ow fear me!) 47 . . . Kc6 48 Bc3 Kd6 49 a is zugzwag, ad game over sce either Back's kg or hs bishop gets defected from protectg his respective paws.

47 fxg4 1-0 ftr 47 . . . Bf (th bishop ds his og, sf-imposd xi ad rturs to socity a dffrt ma  but h fds upo his rtur, that a thos who h ovd ar dad) 48 h4, Whit ow has thr passrs, whras Back's pathtic doubd -paws ar asiy hd.

Chapter Six Botvinnik on Endings Botvik's remarkabe acuity i edigs arose from  cacuatio power ad his itese receptivity to eve the most mute shift  the geometry. Late  his career, whe  midde game powers were o the dece, Botvk bega to switch to tecica positios, where he sti reiged supreme. The game which pops out to me is his opposite-cooured bishops edg versus Kotov (game 60). I his, Botvik sacs two paws, gog from paw up to paw dow ad, impossiby, ws the game. The effect of his doube paw sac hit me ike a geometer's shock at dscoverg a four-sided triage. Game 48 M.Botvinnik-T.V.Petroian Word Champioship (14th matchgame), Moscow 1963 Queen's Gambit Declined 1 d4 d5 2 c4 e6 3 Nc3 Be7 4 cxd5 exd5 5 Bf4 c6 6 e3 Bf5 7 g4 Be6 8 h3 Nf6 9 Bd3 cS 10 Nf3 Nc6 11 Kfl 0-0 12 Kg2 cxd4 13 Nxd4 Nxd4 14 exd4 Nd7 15 Qc2 Nf6 16 f3 Rc8 17 Be5 Bd6! 18 Rael Bxe5 19 Rxe5 g6 20 Qf2 Nd7 21 Re2 Nb6 22 Rhel Nc4 23 Bxc4 Rxc4 24 Rd2 Re8 25 Re3 a6 26 b3 Rc6 27 N a4 b6 28 Nb2 as 29 N d3 f6 30 h4 Bf7 31 Rxe8+ Bxe8 32 Qe3 Bf7 33 g5 Be6 34 Nf4 Bf7 35 Nd3 Be6 36 gxf6 Qxf6 37 Qg5 Qxg5+ 38 hxg5

: Who stads bttr hr?

1. White's kight, with access to e5 ad f4, ooks superior to Back's bshop, who s hdered by his ow d5-paw. The presece of fixed paws o g6 ad h7, the same coour as the bshop, doest hep Back either. 2. White's kg has quick access to the cetre via g3, f4 ad e5. 3. White domates the cetra dark squares, may made possibe by the presece of hs g5-paw. 4. Back cotros the c-fie  ot eough to offset White's three advatages. 38 ... a4? A mstake idea. Back hopes to eimate a queeside paws, but Botvik's  Lt's tay:

respose dodges the attempt ad creates further weakess i Back's camp. 39 bxa4! Much stroger tha 39 Ne5 Rc3 40 bxa4 Ra3 41 Rb2 Rxa4 42 Rxb6 Rxa2+, whe the tradig away of the etre queeside woud be hepfu to Back's cause. 39   Rc4 40 as! Botvik  quick to retur the paw ad seize cotro over c. The weakeg of this square is a fudameta Back must dea with, the way a chid burts out a upeasat truth the aduts i the room refuse to discuss. As usua, Botvk is quick to reject ay e which ws materia but oses the itiative. So he avoids 40 Nb2 Rb4, which ties ite dow. 40  bxa5 41 N cS Bf5 42 Kg3!





Preparg to trave deep to Back's camp. 42  a4 : What is th prpos of this mov?  Ptrosia pas to pay th paw to a3, whr h ay atr hav tricks, i h ca gt his rook to b2 or attack Whit's a-paw with . . . Bb.

43 Kf4 a3 44 Ke5 The kg views d5 with icy aimosity. 44   Rb4! 45 Nd3! Botvik writes: "Of course, with such a active kg, ite ca aow traspositio to a rook edg but shoud ot permit the eemy rook to vade o b2. Not 45 Kxd5? Rb2 ad if 46 Nb3?? Bbl, suddey Back is w g. 45  Rb5 It  too soo to traspose to a rook ad paw edg: 45 ... Bxd3? 46 Rxd3 Ra4 47 Kxd5 is hopeess for Back;  e. g. 47 ... Kf7 (47 ... Ra5+ 48 Ke6 Rxg5 49 d5 is eve worse) 48 Ke5 Ke7 49 d5 Ra5 50 Rb3 Ra7 51 Rb8 (threateg Rh8) 51 ... Kf7 52 Kd6 ad Back ca resig.

 (  ): How dos Whit mak progrss?

 Th kg is a fightg pic", xtod titz. Giv chas to Back's rook with th kg.

46 Kd6! Kf7 47 Kc6! The waderig kg takes o magfied fuece from his ew outpost. 47 000 Bxd3! Petrosia correcty paces his faith  the drawish ature of rook ad paw edgs. The X-factor is ite's super-active kg. 47 . . . Ra5? 48 Nc5 eads to the oss of d5 with a hopeess game for Back. 48 Rxd3 Rb2 49 Rxa3 Rg2 50 Kxd5 Rxg5+ 51 Kc6

: How is this dffrt from th rook dg w ookd at arir?

 I this vrsio Back st drs ay dffictis, but both his kg ad rook ar far mor

activ tha  th hopss arir vrsio w xamd. Back's dffictis hr:

paw.

1. He is dow a

2. Hs passed h-paw s sower tha White's d-paw. 3. Botvk's rovig kig assists his passed paw up the board. The deficiecies i Back's game are aid aked  yet he may sti be abe to draw! Such s the drawg power of rook ad paw edgames. 51   h5 52 d5 Rg2 53 d 6 Rc2+ 54 Kd7





We eter depths difficut to fathom. Desre for promotio to a ew quee is to be the boody area, the pace of strivgs, ife ad death. I the swr of the positio are dua aveues, dua fates. Oe e oses, the other draws. The time has arrived for us to eter our sge, private future.

 ( ): Back ca pay 54 ... h4, rog his

h-paw forward. Or he ca toss

 the preparatory 54 ... g5, which prevets White from payg f3-f4. Choose wisey.

54  h4? For ay machie to ru smoothy, every eemet must cotribute positivey to the whoe. I this positio, this move s the sge jarrg aomay  a otherwise heathy etity. Perhaps Petrosia seeks ratioaity where oe exists. Exuberace is o praiseworthy trait whe the zea s directed  the wrog directio, ad the grace period o upushed errors is about to expre. Now Back's h-paw s the prcipe actor  a bady performed pay, ad the adjacet racers do ot trave at a equa pace . I ths stace, quaity supersedes speed.  54 ... g5! was corrct. Botv cams Whit shoud sti w, but Hdini aaysis ovrrus. I dfdd with Back ad Hdini faid to produc a w for Whit  ay vrsio.  samp : 55 Kd8 h4 56 Ra7+! Kg6 57 d7 h3 58 Ra6+! Kf5 (ot 58 ... Kg7?? 59 K7 Rd2 60 R6! h2 61 R ad ws) 59 h6 h2 60 a4 Kf4 61 a Ra2 62 K7 R2+

·



(Back's rook, a sotope of his twi brother o h6, may ot be as active or fuetia, yet he remas geeticay idetica, therefore fused with equa potetia as hs more charsmatic sibg) 63 Kf6 Rd2 64 Ke6 Re2+ 65 Kf7 Rd2 66 Ke8 Re2+ (Back's aoyg rook cotuay cears his throat to get the white kg's attetio) 67 Kd8 Ra2! 68 Ke7 Re2+ 69 Kf6 Rd2 70 Ke6 (ite's frustrated kg s uabe to cross over the impervious boudary before him  he keeps fdg himse scruched , ike a card tabe, four egs foded, before beg put away) 70 ... Re2+ 71 Kf7 (the white kg's pa at the costat rebuffs may be more psychosomatic tha rea; sti, pa s pa) 71 . . . Rd2 72 Ke8 Re2+ (the rook cotues to foow, a private apparitio of a dead brother, oy see by ite's kg ad vsibe to the rest of humaity) 73 Kd8 Ra2 with a draw, sce 74 a6 s met by 74 ... Rxa6!. 55 f4! At ths pot a deepy troubg covictio bubbes up from the depths of Back's subcoscious. He ow uderstads his pa has goe bady awry. hat was oce rumour ow morphs to reaity: Back is too sow  the race. 55  Rf2 Petrosia retaiates as best as he is abe, but it just s't eough. 56 Kc8!

56  Rxf4 Ateratives fai. For exampe, 56 . . . Rc2+ 57 Kd8 or 56 . . . Ke6 57 d7 Rc2+ 58 Kd8. 57 Ra7+! 1-0 Th rook sits back  his chair ad puts his ft up o th dsk, with a smug ss of accompishmt.

This mov aows Whit to bock a c-f chck with Rc7, so 57 ... K6 58 d7 is curtas. Game 49 MBok-MTal Word Champioship (1st atchgam), Moscow 1961 Nimzo-Indian Defence 1 c4 Nf6 2 Nc3 e6 3 4 Bb4 4 e3 0-0 5 B3 5 6 a3 xc4 7 Bxc4 B6 8 Nf3 Nc6 9 Nb5 e5 10 Nx6 Qx6 11 xe5 Qx1+ 12 Kx1

: How woud you assss this dg?

Answer:

1. White has the bishop pair i a semi-ope positio. 2. Back eads microscopicay  deveopmet. 3. Back, dow a paw for ow, regas it o his very ext move with a doube attack. 4. d3 coud be a source of worry for White, especiay sce wi soo be ready with a e5-kight traed o that square. 5. White's kgside paw majority ad Back's queeside versio seem to baace each other out. 6. White's kg is  the cetre, which for ow is a wash betwee a active kg ad a edagered kg. Cocusio: Chaces ook dyamicay baaced. 12  Ng4 13 Ke2 Ncxe5 14 Bd5 Botvik may have feared the e 14 Nxe5 Nxe5 15 Bb3 b6!, whe Back's bshop arrives with a spash o a6. 14  c6 15 Be4 Be6?! The ever-optimstic Ta may be payg for the w, but it becomes apparet that he misjudged his chaces. He shoud pay 15 ... b6! 16 Nxe5 Nxe5 17 f4 (Back is aso at east equa after 17 f3 Ba6+ 18 Kf2 Rfd8) 17 . . . Ba6+ 18 Kf2 f5 19 Bbl Nd3+ 20 Bxd3 Bxd3 21 Bd2 with a ear certa draw. 16 N d2 Rad8 17 h3 Nf6 18 Bc2 Rd7 19 b3 Seg cotro over c4. 19   Rfd8 20 Rdl Nd3 The kight eaves his ow safe precct to eter White's ecave. There s o rea choice sce White's bishops come aive after 20 . . . Nd5 21 Bb2 Ng6.

21 Bxd3 Rxd3





The white kg's eyebrows rise  digat arcs at the trusio. The power  the positio draws earer d3, ike iro figs to a maget. Yet predictabiity s a iabiity o the chess board ad appearaces deceive. Lookg coser, Back's apparet itiative is spet, ad Ta's positio soo apses to ethargy, despite the opposite dicators of the give data: 1. Back eads  deveopmet. 2. b3 appears to be  dager. 3. White ost his proud bishop par. Whe I show this positio to studets they uiformy prase Back's positio. Botvik had see deeper to the edg's secrets ad,  reaity, Back stads sighty worse ad must tread carefuy.

 (  ): What woud you pay as Whit?

 22 Bb2!

Cofusio upo paradox upo iroy, yet Botvik retas his beargs ad sees through the subteties of the positio. The threat to b3 proves iusory due to a back rak trick; whereas White threates to pick off a exchage with Bd4!, cuttg off the d3-rook from its protectio. Ad asty, White threates to damage Back's structure with Bxf6. Now the positio mysteriousy gets mouded to White's favour, by meas secrety kow to Botvk, oy through hs ubeievaby deep strategic uderstadg. 22  R3d7 Oe hears sury murmurs emaatg from Back's camp. Dissatisfactio with oe's ot  ife is huma ature. Eve the Garde of Ede wast good eough for Adam ad Eve. Ta agrees to a damaged structure after sowy digestg the data. Persoay, I thik Ta shoud have tried the exchage sac 22 ... Ne8!? (ot 22 ... Bxb3?? 23 Nxb3 ad Back cat recapture o b3) 23 Bd4! (the oce sought-after d3square trasforms to a repository of pa for the straded rook) 23 . . . Bxb3 (24 Kxd3 Bxd1 25 Rxd1 c 26 Nf3 cxd4 27 Nxd4 s ubaaced, as White's superior kg positio may be offset by Back's queeside paw majority) 24 ... Rxd2+ 25 Kxd2 Bd5, sce it s the most aggressive ad suited hs stye. After 26 Ke1 ! b6 (26 . . . Bxg2 27 Bxa7 b5 28 a4 aso gives White w g chaces) 27 f3 f5, Back has a paw for the exchage. Okay, if ayoe is gog to w, it wi be White, but I sti prefer this

positio to the drab defesive task haded to Ta  the game.

23 Bxf6 Botvik defaces Back's structure, as if compeed by a stct to vadaize. 23 000 gxf6 24 b4 Bf5?! This move merey catayzes his difficuties ad Ta gets pushed back. Back's best chace to rema active ies  the e 24 . . . f5! 25 Nf3 Bd5 26 Nh4 f4! 27 Rd4 (27 exf4?? oses staty to 27 . . . Bc4+ ad if 28 Ke1 Re8 mate!) 27 . . . fxe3 28 Kxe3 Re8+ 29 Kf4 Rde7 30 f3 Re2. 25 Nb3 B d3+ 26 Ke1 b6 27 Rac1 Be4 28 f3 Rxd1 + 29 Rxd1 Rxd1 + 30 Kxd1 B d5? Back may yet hod the edgame after 30 ... Bf5! 31 e4 Bd7 32 g4! (fixg the weakess o f6) 32 . . . Kf8 33 Kd2 Ke7 34 Kc3 Kd6 35 f4 c 36 bxc5+ bxc5 37 Na5 Kc7 38 Nc4 Bc6, but it sti wot be easy.

 (  ): Back is  srious troub, with his

structure damaged ad White's kg

ahead  the cetrazg race . Come up with a cocrete pa for White to w the game.

 Th kight, as i succumbg to a arbitrary whm, gos chasg aftr back paws. Whit must ur th c-paw forward, aftr which kg ad kight ca gag up ad pick it off, ay bcaus Back's kg ras WOL o g8. oo, th kight's assrtio morphs to trrogatio.

31 Nd4! Threateg to uderme the c-paw with e3-e4 ext. 31 000 cS 32 bxc5 bxc5 33 Nb5! a6 Not 33 . . . a? 34 Nc3 Bc6 (to cover both e4 ad a4) 35 Kd2 ad Back cat do much about the comg Kd3, Kc4 ad Kxc5. 34 Nc7 Bc4 35 Ne8! The feisty kight cotues to be uameabe to reaso ad aoygy reiterates threats. 35 000 f5 36 h4 : What is th purpos of this mov?

 Whit avoids ... Bf tricks.

36   Kf8 37 Nd6 Bf1 The bishop goes through the motios with exasperated acquiescece. After 37 ... Be6 38 Nb7 c4 39 Nc Bc8 40 Kc2, hite's kg soo picks up the stray o c4.

38 g3 Ke7 Or 38 ... Bh3 39 Ke2 Ke7 40 Nb7 c4 41 Nc a 42 Nb7 a4 43 Nc, whe the sicky a4-paw is o more tha exercse for the kight's sarcasm. 39 Nxf5+ Back gets sapped with a surcharge of oe paw. 39   Ke6 40 e4 Ke5 41 Kd2 1-0

Back's resistace imps to dre faity.

: Is't rsigatio prmatur, sc Back c try

41 ... Bh3 ad ca fitrate to d4

whe the kight moves?

 Th troub is Whit has o ttio of movg th kight. H ca pay 42 Kd3! d Back c't tak th kight sc th kg ad paw dg is hopss. Othrwis 42 . . . Bf + 43 K3 f6 44 f4+ K6 45 g4 ws, sc Back's c-paw wasts away to atrophid disus, whi Whit crats two passrs o th othr sid. Game 50 MBotvk-VSmyslov Word Champioship (6th matchgam), Moscow 1958 King's Indian Defence 1 c4 g6 2 e4 Bg7 3 4 6 4 Nc3 a6 5 Be3 Nf6 6 f3 c6 7 B3 b5 8 Q2 bxc4 9 Bxc4 5 10 Bb3 xe4 11 Nxe4 0-0 12 Ne2 a5 13 0-0 a4 14 Bc4 Nb7 15 Rac1 Rb8 16 Nxf6+ Bx 17 Nc3 N 18 Be2 Be6 19 Rf1 Bg7 20 Bh6 Bxh6 21 Qxh6  22 R2 B 23 h4 Q 7 24 a3 Rf8 25 Ne4 Qe8 26 B B5 27 Nc5 Qf8 28 Qxf8+ Kxf8

29 Na6 Rbc8 30 Nb4 Bb3?!

: Why did Back had ovr a paw?

 I'm ot so sur about this dcisio ithr, which is dfficut to vaidat mpricay. It ooks to m ik  ovrractio which acks coctio to th actua rqurmts of th positio. mysov bts o activity ovr atria. I this cas h rjcts th passiv  30 . . . Rd6 31 Rdc2 Rb8 32 Rc (32 Nxc6? fas to w a paw aftr 32 ... Bxc6 33 xc6 xd4) 32 ... Kg7 33 Ra Rb7 34 Ra6 Kf7, whr h is admittdy tid dow, but I do't s a way to ak furthr progrss for Whit.

31 Rxc6 Rxc6 32 Nxc6 Rd6

 (  ): For ow, Back's sg qusid paw

hods back ite's two. How

woud you proceed with ite?

 Go aftr a4  two stps:

Step 1: Trasfer the kight to c, where it disturbs the

rest of Back's bishop ad hits a4 as we. Step 2: Pay for Bb5, addig a secod attacker to a4. 33 Na5! Ba2 34 Nb7! Rd5 35 Nc5 ite iteds to foow with Bb5, as we as Kf2-e3, reforcig his d-paw. 35 000 e5!

: Is't Back just takg ovr th itiativ ow?

 Lt's do a xrcis to fd out:  ( ): mysov compicats, raizg

waitg tactics are doomed to

faiure . He threates to uderme ite's ow ustabe kight ad adds pressure to the ped d4-paw. ite doest have time for his teded Bb5. hat shoud he do?

 Doub attack, thratg th f6-paw d a fork o c3.

36 Ne4! Rxd4 37 Rxd4 exd4 38 Nxf6 Ke7!? Smysov, sesg he is gettg the worst of the argumet, raies by gambg, throwg himsef to the fight with his atagoist with reewed vigour. The kg hazards a desperate fight to cetraizatio at the cost of aother paw. Oce aga, Smysov trades materia for dyamism, skg more of his hard-eared moey to the vague, risky eterprse of creased activity. Yet I cat hep but fee his vestmet rus couter to the actua requremets of the positio. Perhaps he had better drawg chaces simpy protectg his paw with 38 . . . h6.

39 Nxh7 Bb1 40 Ba6?! Iaccurate. White shoud pay 40 Bb5!, tyig the back kight to a4. 40   Nd5 41 Kf2 Ne3! Threateg a fork o d1.

42 Be2 Ke6 : Dos Back hav a combatio hr with 42 ... Nxg2 43 Kxg2 d3  ?

 Th combatio mry rgas th pic but th rsutg dg ftr 44 Bxd3 Bxd3 45 Kg3 is totay hopss for Back, who rmas two paws dow ad whos paws ar fixd o th sam coour as his rag bishop.

43 Ng5+ Kd5





The eergy differetia betwee the two armies provides a jarrig juxtapositio. Back's forces a await  optima posts, despite a two paw deficit. Coversey, White is two paws up, but his fame of costructive pay cotues  the midst of pety: 1. He s tied dow by Back's passed d-paw. 2. Back threates to make it to a race with . . . Nc4. 3. Back's kg puts hs active couterpart o f2 to shame, who ooks as ucomfortabe as I do whe awaitg my doctor's etry  his office, sittg o a tissue-covered examatio tabe  my uderwear.

4. White's kight has o targets. 5. White has dfficuties  activatg paw majorities o either side of the board.

 ( ): Dspit a ths dfficutis,

Botvik discovers a

aomay, a oophoe  covetioa ogic, which aows him to eradicate Back's gas by tearg the fabric of the positio ad stitchg it together aga to his ow ikg. How did he do it?

A By offrg to r tur o paw. Th cratur from aothr dimsio is ft bfor it is s. Back's positio rachs its sharpst pak ad has owhr to go but dow. 44 Ne4!! Bxe4

Nobody wats admittace to aother's misery, so the bishop caousy bai out of the game. Decg with 44 ... Bc2 45 Nd2 is competey hopeess as we, sce White's kight becomes the perfect bockader o d2. 45 fxe4+ Kxe4 46 g4! Threateg h4-h5!. The cerk assumes a busy, preoccupied attitude whe the boss passes by. 46  Kf4 The immediate 46 ... d3 is equay hopeess: 47 Bf3+ (the bishop decides to straighte out Back's deusioa pieces o matters of obediece toward their betters) 47 . . . Kf4 48 h5! (ow truth ufods before the dazed back kg's eyes: the h paw creeps away, ke a arceous buter who makes off with the master's best siverware  the defectio power heret  the passed h-paw costs Back the game) 48 . . . gxh5 49 gxh5 Nf5 50 Bdl (the bishop, quick as thought, trasfers to dl ad eyes a4 with decisive effect) 50 ... Kg5 51 Bxa4 Nd6 52 Bb3 (coverg agast ... Nc4) 52 ... Kxh5 (Back's kg is shor of a power ad excommuicated to a wastead as peace) 53 Ke3 ad White ws. 47 h5! gxh5 48 gxh5 Kg5 48 ... Nf5 49 Bdl does the job too. 49 Kf3! 1-0

Each momet White's kg is aowed to grow  power  a momet of dmishg potetia for Back's kg. Smysov decides to avoid proogg resistace  a futie cause . 49 . . . Kxh5 50 Ke4+ Kg5 51 Kxd4 Kf4 52 Bb5 picks off the a4-paw.

: Tru ough, but what i Back gos ftr b2?

 Botv's fa, hidd pot: th kight gts trappd aftr 52 . . . Nd1 53 Bxa4 Nxb2 54 Bb3!  ad vry mportaty, Whit has th corrct coour bishop for his a-paw. Game 5 Bok-AAlekhe VRO Touramt, Th Nthrads 1938 emi- Tarrasch Defence 1 Nf3 5 2 4 Nf6 3 c4 e6 4 Nc3 c5 5 cx5 Nx5 6 e3 Nc6 7 Bc4 cx4 8 ex4 Be7 9 0-0 0-0 10 Re1 b6 11 Nx5 ex5 12 Bb5 B7 13 Qa4 Nb8 14 Bf4 Bxb5 15 Qxb5 a6 16 Qa4 B6 17 Bx6 Qx6 18 Rac1 Ra7 19 Qc2 Re7 20 Rxe7 Qxe7 21 Qc7 Qxc7 22 Rxc7

: Is Back osg du to Whit's rook cotrog th svth rak?

 Back may dd b osg at this pot, but it taks rmarkab tchiqu to covrt agast a dtrd Word Champio. kh forsaw a way to chag o th svth rak with his xt mov. I woud say th positio havy favours Whit, but a actua w ras  th distac.

22 000 f6! 23 Kf1 If 23 Rb7 Rc8 24 Kf1 b5 25 Rb6, itedig Rd6, the 25 . . . Kf7 hods the positio together, sice 26 Rd6? is met by 26 ... Ke7! 27 Rxd5 Rc2. 23 000 Rf7 24 Rc8+ Rf8 25 Rc3 The most active square, where the rook may trasfer across the third rak  eeded. The rebuffed rook fais to make progress aog the eighth or seveth raks ad must agree to a tactica retreat for the momet. For ow, Aekhe erects a seemgy impregabe barrier to the rook's attempts at peetratio. 25 000 g5!?

 (  ): Typica kh, who opts for th most

aggressive defesive posture,

eve  it meas rskig further weakeg. How woud you go about stregtheg ite's positio?

 Trsfr th kight to 3, via c2. From thr, th kight prssurs d. Whit aso has th stratgic thrat g2-g4, foowd by Nf, strgthg th kight's positio furthr.

26 Nel! h5!? It might be argued that Aekhe cotiues to bow upo the embers of a fire which burs him. Optimsm kows o bouds or fears peri. Oce agai Back opts for aggressio at the cost of potetia overextesio. Aekhe esures hite cat foow with the g2-g4, Nf5 pa ater o. Our 25-poud ruty terrier, Kahess, aways wofs dow the distributed doggie treat before the other dogs. Step two occurs whe Kahess rather optimisticay approaches Nikki, our l-poud behemoth of a Akita, hopig to puro her doggie treat. Statisticay, the resuts rarey favour Kahess  these tiresome cae ega dispute s. I this game Aekhe s Kahess-ike optimism receives the sharp rebuke of Botvik's paws . I'm ot sure if this is a case of a payer cojog optimsm with msassessmet (givg birth to disaster) or just a good practica try  a hopeess strugge!

: houd't Back's ast mov b cosidrd a rror

sce he seems he bet o

creatg fresh weakesses?

 Back is virtuay  zugzwag. If h tris to activat his kg with 26 ... Kf7, Whit ras  cotro aftr 27 Rc7+ Kg6 28 Nc2 h 29 K2 R8+ 30 Kd3 R6 31 Nb4 Rd6 32 Rc8 a 33 Nxd, wg a paw sc 33 ... Nc6?? is mt by 34 xc6!, wg a scod paw.

27 h4!!

·



Immediatey sezig upo Back's ast move. Back's provocatios oy resut  atagoizig a dagerous foe who ow seeks reprsa. Aekhe soo ears a simpe fact which has stragey euded poitica eaders throughout mieia: war is expesive. T carefuy about decarg it if your coutry is broke to beg with.

: What was th purpos of Whit's ast mov?

 Whit sks to crat a statioary targt o h5 ad aso duc ovrxtsio.

27   Nd7 Istead: a) 27 . . . Kf7 28 Rc7+ Ke6 29 hxg5 fxg5 30 Rh7 h4 31 Nf3 ws a paw, sce 31 . . . Rg8 s met by 32 Rh6+, pickg off b6. b) 27 . . . gxh4 28 Nd3 Kh7 29 Nf4 Kh6 30 Rh3 Rc8 31 Rxh4 Rcl + 32 Ke2 Rc2+ 33 Ke3 Rxb2 34 Rxh5+ Kg7 35 Rxd5 Rxa2 36 Rd6 Rb2 37 Nd5 ad at east oe back paw fas,  ot more. 28 Rc7 R£7 29 N£3! The creature s roused from its ar    29   g4 30 Nel . . . oy to eter its dark recesses oce aga. Itetio: Nd3-f4.

: It ooks to m ik Botvk dugs 

too may revisios of his pa. Do you

agree?

 I disagr. Too may rvisios of o's ps do dd hav a way of smudgg fficicy, th way too ay rasurs o th sam spot abrad ad fray a sht of papr. But this stac is ot a xamp of myopic atttio to dta ro dg a p's atura fow. Botvik's push-ups with his kight ach srv a purpos ad ach  tur xtract cocssio from his dsprat oppot.

30  £5 31 N d3 £4 A massive fightg force sprts forth, deeper to trackess, hostie terra, dested to trasform to a boodied, haggard skeeto bad, devoid of stregth or morae. Back is ow officiay overexteded, but to aow Nf4 s uthkabe. 32 £3!

·



The byways of Back's crcuatory system remai cogged ad his pseudo-attack draws the sow, aboured breaths of the deathy i. As we have see throughout the book, Botvik habituay rejects materia i favour of the itiative. I ths case he creates a target o f4, rather tha be coaxed ito the ie 32 Nb4 Nf6 33 Nxa6. 32 000 gxf3 33 gxf3 as Prevetig Nb4.

34 a4 Fixg the paws o the other wg. 34 000 Kf8 35 Rc6 Ke7 36 Kf2 RfS 37 b3 Kd8 38 Ke2 Nb8! A great practica try.

39 Rg6! The rook, a creature of the heaves, refuses to debase himsef with earthy sorrows of ageg, sickess ad death. He fits about, ike a fy who just sits o the wa for o reaso, ad the just as radomy fies off to owhere  particuar. The text tex t s mu much ch stroger tha grabb grabbg g b6, as Rg5 is ow  the air. air . Botv Bo tvk refuses refu ses to cocer him himse se  with trif trifes es ad ad rejects reject s the paw offer.

: Why o arth did kh had ovr th b-paw,

ad the, eve more perpexg,

why didt Botvk accept?

A A Back's ast mov is a dspr d sprad adoo . kh kh is wg to pay a stp st p pric to xhum xhum his og brid kight. kight. 39 xb6? is mt by by 39 . . . Kc7 40 Rg6 Nc6, Nc6, wh Back rgas th th ost paw ad grat grats s a sorts of cotrpay o th qusid.

39  Kc7 Kc7 40 Ne Ne5! 5! Preve Prevettig both both . . . Nc Nc66 ad ad . . . Nd7. 40  Na6 Na6 Back's kight, fay uecumbered of its burde, skips ad bouds out with regret for precious paytime ost. The kight, which oce ay ert o b8, imperceptiby twitches hs facia musces ad hs eyes uexpectedy ope. Here the game was adjoued. 41 Rg7+ Aekhe tod Botvk he woud resig if he had seaed 41 Rg5!, which, as it turs out, s the Houdini approved move as we. 41   Kc8

 ( ): Fd a forcg  for Whit which ws a paw.

A A Fork For k doub do ub  attack.

42 Nc6! Threateg Ne7+. The mysterious kight cotues to brg uwecome tidgs. Oe ca't ca't fai fai to be dazzed by Botvk' Botvk'ss hypot hypotic ic  amost amost soporific soporifica ay y dsorie dso riet tg g  kight kight gyrat gyratios. ios. 42  Rf6 Rf6 The cectio betwee Back's rook ad d-paw s severed, as ite's ast move dsturbed the oce happy coupe's coubia biss. Regrettaby, the rook must abado his comrade o d5 due to the fork threat o e7. 43 Ne7+ Kb8 44 Nxd5 Nxd5 Success aight aightss upo upo  ite's baer. baer. 44  Rd6 45 Rg5! Dua purpose purpos e : Coverg d5 whi whiee meac meacg g h5 h5 . 45   Nb4 46 Nxb Nxb44 axb4 axb4 At og ast, Back rid himsef of his pariah kight, but this fact hardy esses Aekhe khe s misgivgs misgivgs sce  pos p ositio itio degrade de gradess to t o a eve eve  the oppo o pposite site of

reassurg, as he  duce duced d to a hopeessy hopee ssy ost rook roo k ad ad paw edg. edg.

47 Rx5 Rc6 47 ... Rxd4 48 Rf5! threates the advace of the h-paw. If Back's paraysed rook moves he drops dro ps more paws. paws. 48 Rb5!





Remov Removg g ay possibe possibe . . . Rc3 ad ad . . . Rxb3 Rxb3 couter couterpay. pay. 48   Kc7 49 Rxb Rxb44 Rh6 Rh6 50 Rb5 Rxh4 Rxh4 51 Kd3 Kd3 1-0 kh said this gam was th oy o  th grat VRO touramt whr h ft compty outpayd by his oppot. I this vt, Botvk achivd th mpossib by dfatg both Capabaca ad kh, producg mastrpics  ach gam. It woud b a fu t yars bfor Botvk bcam word champio hmsf, yt h, ik Magus Cars, showd that you do't cssary hav to b word champio to b rcogizd as th bst payr  th word. Game  PKereMBotvk Chigor Mmoria, Moscow 1947 Dutch Dence 1 4 e6 2 Nf3 f5 3 g3 Nf6 4 Bg2 Be7 5 0-0 0-0 6 4 4 5 7 N3 6 6 8 Rb1 Kh8 9 x5 x5 10 Bf4 Bf4 N6 11 Ne5 B7 B 7 12 Rc1 Rc1 R8 13 Q3 Nh5 14 B2 B6 15 Nx6 Bx6 Bx6 16 Qf3 Qf3 Qe8 17 Q3 Nf6 Nf6 18 a3 R7 R7 19 Bg5 Ng4 20 Q2 N 21 Bf4 Q7 22 Bx6 Qx6 23 Qf4 Qxf4 24 gxf4

: Woud you say Whit stads a shad bttr du to Back's bad bishop?

 I this stac, I woud rat chacs at v sc Back's bishop ca b rpostd to h, or v b ad a4 aftr Back doubs d oubs rooks roo ks o th c-fi c-fi. . Furthrmor, Whit's bishop, though tchic tchica ay y  goo d", d" , is't ray so good aftr a, sc it hits a paw wa o d.

24 000 Rfc8 25 e3? If we mess up our ife through a uwise decsio, we rarey get a secod chace to fix the se-ficted woud. We oy get oe ife, ot two, so it's importat to get it right the frst time. Keres vioates the prste ature of the positio, the way a oi spi does to a ecoogicay sesitive area. This s oe of those automatic moves which ooks fe  the abstract, but upo deeper examatio, s foud to be ackg. What is remarkabe to me is how such a ocuous move cost Keres the game  what appears to be a competey uosabe positio. What I eared from this game is the maddeg epiphay: o matter how much I study chess, there s aways so much more which I dot uderstad. Rather tha go for this worthess accoutremet, White shoud opt for 25 Na2! Bb5 26 Rxc7 Rxc7 27 Rcl! Rcl ! Rxcl+ 28 Nxcl, mata mata g g the baa baace . 2 5 000 Bb5! Bb 5! 26 2 6 Rfel Kg8 Kg8

2 7 £3 : Why Why ot chag chag th b-bishop with 27 Bf - ?

 Back ftrats  th  27 ... Bxf1 28 xf1 N4 29 Na2 Rc2. Now i 30 b3? R8c6 31 xc2 xc2 32 Nb4 Rc3 33 Rb1 a 34 Na2 Rc2 ws.

27 000 Bc4! Perhaps thi thik kg g about abou t . . . Ne8-d6. Ne8- d6. Back's cotro over c4 gives gives him him a ty ty edge . 28 Bfl Ne8! The kight's services are eeded o the queeside, Back's theatre of operatios. 29 Bxc4 Rxc4 30 Kf2 Nd6 31 Ke2 b5 Now . . . b5-b b5-b4, 4, or frst frst . . . a7-a5, s   the air. air. 32 Kd3 b4! 33 N a2!?

: Is't this a cas of cam givg way

to aarm, sice ite ow just oses a paw?

 Krs sacs a paw whr th bood of th mk woud curd  far. By dog so h avas himsf himsf of a chac to to cras th v of ffici fficicy cy  his his positio po sitio  which which  th d h vr vr gts. It's It' s asy to swap o vic for aothr. I kow a guy who bg to chw icot gm to quit smokg. Th troub was h got addictd to th gm! Krs' sac dos appar to ook k a cas of a prso attmptg to cast a vot th day aftr th ctio. I th d, aftr mticuous pag ad xps, Whit's podrous projct is dstd to av hm disappotd, sc his fictioa pic activity wofuy fas to compsat.

Sti, it's difficut to faut Keres sice he woud remai bady tied dow  the e 33 axb4 Rxb4 34 Rc2 Nc4 35 Ra1 a 36 Nd1 Rcb8 37 Ra2, whe ite ca barey move. Back, o the other had, ca brg up his kg ad expad o the kgside, ookg for targets. 33  bxa3 34 bxa3 Ra4 The a3-paw is doomed. 35 Rxc8+ Nxc8 36 Nc3 Rxa3 37 Kc2 Nd6 Coverg b7 fi fitrati tratio o attempts. attemp ts.

38 Rb1 Kf7 39 Rb4 39 Rb8 ooks sighty more active. 39  Ral! Ral! Targets: e3, f3 ad h.

4 0 K d3 : Why ot pay 40 Rb



?

 Krs dspraty hops to avoid a kight dg, th scod worst dg to b dow a paw (bhd a kg ad paw dgam, which is gray th worst cas scario for th paw dow sid).

40  Ra3! Ra3!?? 40 . . . Rh1 Rh1 ca be met by 41 Ra4 Ra4 Nc8 42 Ra2. Ra2.

41 Kc2 Ra1

Trademark Botvk repetitios, to draw coser to the time cotro ad try to extract errors from the defeder.

42 K d 3

 (  ): ): Back has a choic of thr pottia soft spots  Whit's

positio: e3, f3 ad

h2. O which of the three shoud he focus  attetio?

  dark cmus forms ovr 3, th most tdr spot  Whit's positio.

42  Re! Re! Threateg ... Nc4.

43 Ra4 43 Ne2? a!  hopeessy ost for ite as we. 43   Nc4 44 Rxa7+ Rxa7+ Kg6 Kg6 45 e4 ite's oce oce tight cetra formatio formatio has bee ured forward forward to t o its doom. do om. 45  Re3+ Re3+ The suppe rook twists ad weaves its way ito the uderbey of ite's paw mass, which ow takes o the disheveed ook of a frat house the Suday mog after a party. 46 Kc2 Rxf3 47 exf5+ 47 exd5 is met by 47 ... Ne3+ 48 Kd2 Nxd5 49 Nxd5 exd5 50 Ra6+ Kh5 51 Rd6 Rxf4 52 Rxd5 g5 53 Rd7 Kg4 54 d5 h5, whe Back shoud w the queeg race. 47  Kxf5! 48 Rxg7 Rf2+ 49 Kb3 49 Kd3?? Rd2 mate woud be a embarrassg ed to the game. 49  Rb2+ Rb2+ Back's rook ad kight team prove to be the architect of the white kg's peptic stress eve.

50 Ka4 Rx2

ite cat hod the edig: 1. He is dow a paw. 2. He urses a par of weak we ak g g isoais. isoais. 3. Hs kg is cut off o the west coast whie Back's remas active ad vibrat. 51 Rf7+ Kg6 52 Rf8 Nd6! Target: d4, via f5. 53 NS Nf5 54 Nc7 Re2 55 Ne8 Nxd4 Reetess tecique. Paw umber two fa. 56 Rf6+ Kh5 57 Rf7 Nf5 Botvik is happy to trade  h-paw for hite's f-paw, which woud aso occur  the e 57 ... h6 58 Rf6 Kg4 59 Rxh6 Kxf4. 58 Rxh7+ Kg4 59 Rd7 Kxf4





The package arrives o f4, which Back's kg eagery uwraps.

: Why dos Krs pay o?

 Whit chrishs chrishs o fa hop: to sac his kight for Back' s two ra r ag g paws, paws, aftr which h woud hav xct chacs to draw. Botv, of cours, is much too prcis to aow this to happ. 60 Nc7 Ke5 61 K Rc2!

Back's rook, havg itte toerace for subordatio, sees to it that hite's kg

participates i the festivities as itte as possibe. I fact, Botvk's excruciatigy sow, carefu techique borders o cruety. 62 Kb3 Nd4+ other swig of the bade cuts the white kig's aready scarred face. 63 Kb4 Rc4+ 64 Ka5 The kig, rug for his ife, takig itte terest  ayoe's troubes but hs ow. 64  Nf5 65 Kb6 d4 66 Na6 Nd6 67 Nc5 Kd5 68 Nd3 e5 69 Rh7 Rc6+ 70 Ka5 Nc4+ 71 Kb5 Rb6+ Back to the guag.

72 Ka4





The kg grudggy agrees, whie mutterg vie obsceities uder hs breath. 72  Nb2+ 73 Ka5 Of course a exchage of kights meas stat resigatio for ite. 73  Nc4+ 74 Ka4 Rb8 75 Nb4+ Ke6 76 Nc6 Nb2+ 77 Ka3 Nc4+ 78 Ka4 Rb1 79 Rh6+ Kf5 80 Nb4 0-1 Krs has had ough ad cocds th futiity of furthr pay. Game 53 PKeres-MBotvk Word Champioship Touramt, Th HaguMoscow 1 rench Defence 1 4 e6 2 e4 5 3 N2 c5 4 ex5 ex5 5 Ngf3 a6 6 xc5 Bxc5 7 Nb3 Ba7 8 Bg5 Nf6 9 Nf4 0-0 10 Be2 Q6 11 0-0 Ne4 12 Be3 Nc6 13 Nxc6 Bxe3 14 fxe3 bxc6 15 B3 N 16 Qe1 Ng4 17 Qh4  18 Rf4 Ne5 19 Qg3 Ra7 20 Ra Ra 21 N4 Nx3 22 cx3 c5 23 Nf3 Qb6 24 Rh4 h6 25 Ne5 Rf6 26 4 cx4 27 Rx4 Qxb2 28 Rx5 Be6 29 R4 Kh7 30 N7 Bx7 31 Rx7 Rg6 32 Qf3 Qe5 33 R4 Rb8 34 Qf4 Qe6 35 R2 Rb5 36 h3 Re5 37 Kh2 Rf6 38 Rf1 Re4 39 Qb8 Rxe3 40 R8 Qe5+ 41 Qxe5 Rxe5

: What is yor stmatio of Back's wg chacs?

 Cosidraby ss tha 50% . Rook ad paw dgs a paw up ar hard to w. Back crass his wg chacs, howvr, by th prsc of th par of a-paws o th board.

42 Rld2 g5?! Premature. Keres takes immediate advatage. 43 g4! A cug itte move ad Keres' famiiar sigia: acute awareess of hidde matig patters. A straight ie sometimes ooks out of pace i a uiverse of crooked ages. 43  Rf7 : Why dos't Back s th

opportuity to create a passed f-paw with 43 . . . f4

-?

 You wakd to th trap!  2d7+ Kg6 45 Rg8 is at! Back's kg, possiby th uckist chss pic vr, arvs at th ft vagaris of fat.

44 R8d7!

Principle Rook versus rook is usually easier for the defending side than a four-rook ending.

44  Kg7 45 gxf5! Principle The defending side in a rook and pawn ending should reduce the nu mber ofpawns  possible.

45  Rexf5 46 a3 46 Rxf7+ ! Kxf7 47 h4!  much simper, reducg the umber of paws further, sce 47 ... g4? is met by 48 Rd6. 46   Rf2+ 47 Kg3 Rxd7 47 . . . R2f3+ 48 Kg4 Rxa3 49 Rxf7+ Kxf7 50 Rd7+ Ke6 51 Rh7 Ra4+ 52 Kg3 shoud be draw as we. 48 Rxd7+ Rf7 49 Rd4 Rf6 50 a4 Kg6 51 h4! The same prcipe of paw reductio foowed. 51  Kh5 52 hxg5 hxg5

 (  ): Com up with a car

pa which eabes White to hod the draw.

53 Rd3? A serious error  a Bad-Aid ist medice. Up uti ow, it has bee Keres makg a the strog moves, but here he vioates a key prcipe: Always remain active in rook and pawn endin gs. I fact, it is ofte better to give away a paw, rather tha aow your pieces to assume a passive attitude. A 53 Rd5! was corrct, targtg a. Th prcips ow th ky to ockg th igma, as I w r-stat: Always remain active in rook endings! ftr 53 . . . Rc6 54 Ra5! (tyg Back dow to his a-paw  h must toss th paw i h is to mak progrss) 54 ... Rc3+ 55 Kg2 4 (55 ... Rc6 56 Kg3 is o bttr) 56 xa6 Rc2+ 57 Kf1 Kg3 58 a f2+ 59 Kg1 Ra2 60 Kf1 g4 61 Ra8 Ra1 + 62 K2 Kg2 63 a6 g3 64 a7, th po sitio is draw.

53  Rf4! Now White's rook is reegated to a passve posture, ad Botvik beds the positio to his wi. 54 Ra3 a5 55 Kh3 Rb4 56 Kg3 Rf4 57 Ral Rg4+ 58 Kh3 Re4 59 Ra3

 (  ): Krs bakd o this positio, corrcty

thkg Back was uabe to

make progress. He was mistake. How did Botvk seize upo his best wg chace?

A tp 1: Back's kg hads to th qusid.

59  Kg6! Botvik s ufooed, the way a patiet seses artifice  the too-soothig voice of hs psychiatrist. ite's thi veeer of security fas, oy to revea a ew reaity: Keres' attempted fortress is a ufouded pa. 60 Kg3 Kf5 61 Kf3 Ke5 62 Kg3 Rd4 63 Ra1 Kd5 64 Rb1 Rb4 65 Rf1 Ke5 Not yet 65 ... Rxa4?, whe 66 Rf5+ Kd4 (or aywhere) 67 Rxg5 Ral 68 K s a simpe draw for ite, who keeps the opposg rook tied to the a-paw, whie the back kg caot approach because of rook checks from the side . 66 Re1+ Kd4! Step 2: Abado the g-paw ad cocetrate o promotg  soo-to-be passed a-paw.

67 K2 Now if 67 Kg4 Rxa4 68 Kxg5 Rc4, Back ws easiy because the white rook ad kg are mispaced. 67  Rxa4 68 Rg1 Rc4 69 Rxg5 a4 70 Kg2 Compared with the e after 65 . . . Rxa4? above, Back's rook is o c4 rather tha a, which meas the side checks o oger work. Meawhie ite's kg is a miio mies away, ost  thought.

70 000 Kc3 71 Kf3 a3 72 RaS Kb3 0-1 Krs rsigd, sg th  72 ... Kb3 73 K2 a2 74 Kd2 Ra4 etc. Game 54 MBok-PKeres Word Champioship Touramt, Th HaguMoscow 1948 altic Defence 1 4 5 2 Nf3 Bf5 3 c4 e6 4 cx5 ex5 5 Qb3 Nc6 6 Bg5 Be7 7 Bxe7 Ngxe7 8 e3 Q6 9 Nb2 0-0 10 Rc1 a5 11 a3 Rfc8 12 B3 a4 13 Qc2 Bx3 14 Qx3 N8 15 0-0 Ne6 16 Rc3 b5 17 Qc2 Rcb8 18 Ne1 Nc8 19 Rc6 Qe7 20 N 3 Nb6 21 Nb4 R8 22 Q R6 23 Rfc1 Rxc6 24 Rxc6 R8

 ( ): Not a combatios ad to mat. Do you

see Botvik's itte trick

which eared him the superior paw structure?  25 Rxb6! cxb6 26 Nc6

The pot. White regas the exchage after havg damaged Back's queeside. 26 000 Qc7 27 N xd8 Qxd8 28 Qc2 Qc7 29 Qxc7 N xc7

Back is  troube: 1. His queeside three to two paw majority is uabe to create a passed paw, whie White hods a heathy five to four majority o the kgside. 2. d5 is isoated ad a potetia target, as is b5. The odd eye sores   Back's camp trigger the visua effect of a we-dressed busesswoma wakg to a meetg wearg oe of those joke Groucho Marx pastic eyebrow/gasses/moustache deas. 3. Back must be o costat ookout for a white kg vasio via b4. 4. Botvik himse oce wrote: "Kight edgs are very simiar to paw edgs, ad  both cases a outside passed paw gives good wg chaces. Whe White activates  kgside paw majority, he wi have  passed paw. 30 Nb1!

: Why a votary rtrat?

 O c3, Whit's kight kps a y o both back paw waksss, at b ad d. This  tr tis at ast o back pic dow to th dfc of ths paws.

30 000 Kf8 31 Kf1 Ke7 32 Ke2 Kd6 33 Kd3 Kc6 34 Nc3 N e8!

 (  ): Krs' ast mov pas th trasfr of his kight to c4, whr

it ties ite

dow to b2. How did Botvik rui this tet with oe simpe idea?

 Th thratd Nb4+ ks Back's pa.

35 Na2! £6 35 . . . Nd6?? 36 Nb4+ Kd7 37 Nxd5 picks off a critica paw.

36 £3 Keepg optios of e3-e4 ope. 36   Nc7 37 Nb4+ Kd6 38 e4 Botvik frees his majority. 38  dxe4+

: It sms to m that Back's ast mov hpd Whit out. Rathr

tha egage  a

maer of odd versios ad perpexg sophisticatios, why ot stad hs groud with 38 . . . Ke6 39 Nxd5 Nxd5 40 exd5+ Kxd5 - ?





Doest his kg positio compesate for ite's passed d-paw?  Lt's ook. t this pot, Hdini asssss th positio at qa  ad dd ftr 41 g4 (or 41 Kc3 f5! 42 K4 Kxd4 43 Kxb5 f4! 44 Kxb6 K3 45 K5 Kf2 46 Kxa4 Kxg2 47 b4 Kxf3 48 b5 Kg2 tc) 41 . . . g6 42 Kc3 f5 43 Kd3 h5 44 h3 h4 45 Kc3 f4! (but ot 45 . . . fxg4?? 46 fxg4 K4 47 b3! Kd5 48 g5 Kd6 49 b4! K6 50 Kd3 Kf5 51 K3 Kxg5 52 K4! Kf6 53 Kf4! ad Whit vtuay ws by a sg tmpo, fufg th ida of vry dustriaist to axiiz productio with d ffort) 46 g5 K6, Whit caot mak progrss, sc 47 K4 Kd5 48 Kxb5?? Kxd4 49 Kxb6 K3 50 Ka5 Kxf3 51 Kxa4 Kg3 v ws for Back.

: That sms quit compicatd! You'r ot caig

that both Botvik ad Keres

worked it a out, are you?

 I do't mak that caim. Kg ad paw dgs ar prhaps th most dfficut of a to figr out at th board. o it is bst ot to tr o ss you ar absouty positiv about th d rsut.

As it happes, ite is ot obiged to eter the kg ad paw edgame just yet. Istead, he ca seek to improve hs kgside positio, eavg the edg as a

potetia threat if Back cotues to pass. For exampe: 39 g4! Kd6 40 h4 Ke6 41 Ke3 Kd6 42 g5!





42 ... Ke6 (or 42 ... fxg5 43 hxg5 h5 44 gxh6 gxh6 45 Nd3) 43 gxf6 gxf6 44 Kf4 h5 45 e5! Kf7 46 Kf5 Ne6 47 Nxd5 Nxd4+ 48 Ke4 fxe5 49 f4! Ne2 50 fxe5 ad White is wg sce he is effectivey a soid paw up, ad ow with the superior kg positio too. Rather tha makg a probaby futie attempt to mata the status quo, Keres perhaps fet it was easier to resove the tesio  the cetre straight away.

39 fxe4 N e6 40 Ke3 Watchg for the cheapo o f4. 40   Nc7 41 Kd3 Ne6 42 Nd5 Kc6 43 h4 Weapoed ambiguity. Botvik decides to ga groud o the kgside, a the whie cevery shiedg tet. 43   Nd8 44 Nf4!

Now Nh is  the air. 44  Kd6 :  ot 44



g6  ?

A Far of pshmt provs a powrfu dtrrt. ftr 45 h5! Back suddy oss a paw, o mattr how h rspods.

45 N The assassi eters the target's home, treadig soudessy. 45  Ne6





Back's kight has bee remaded to e6, i defece of g7, ad hed there uder quaratie, where he vets grievaces upo deaf ears. Now the od, ugy wapaper begs to show its face through the ew coat of pait  Back's house .

 ( ): hit has a itt

trick which ws a paw. Ca you fd it?

A idstp Back's kight chck o c, whi covrg th f4-squar.

46 Ke3! Now there  o good aswer to d4-d5. 46   Ke7 47 d5 Nc5 Back's efficiet kight caws the air, swattg at othig. As each move passes, so creases Back's utter sese of poweressess. The two sides are a study  efficiecy cotrasts. 47 ... g6? 48 Nxf6 Kxf6 49 dxe6 Kxe6 50 Kd4 is totay hopeess for Back, sce he  essetiay a passed paw dow  a kg ad paw edg, despite mathematica equaity.

48 Nx The sum of Back's woes compact to a dese ba o g7. Is it ecessary to iterate the pafuy obvious fact that Back is hopeessy busted? Not oy did hite w a paw, Back remas with his cripped queeside, so he  ow effectivey two paws dow. 48  Kd6 49 Ne6!

49  Nd7 As metioed before, the kg ad paw edg  hopeessy ost for Back.

50 Kd4 Seg kg positio. 50   Ne5 51 Ng7 Nc4 52 Nf5+ Kc7 The ervous kg backs up a step as White's kight approaches with boodshot eyes ad bugg ves o forehead.

53 Kc3 Keres forces a mor cocessio, which  o way aters the outcome . 53  Kd7 Or 53 ... Ne5 54 Kb4 Nc4 55 Kxb5 Nxb2 56 Ne3! Nd3 57 Ng4 ad Back ca resig, sce his adveture met with abject faiure.

 (  ): Now w obsrv th dicat wrappg of th doumt

our eyes . Come up with a cear coversio pa for White.

ufodg before

 Pay g2-g4-g5. ftr th xchag of paws, Whit crats two dady ctra passrs which Back ca't hat.

54 g4! Ne5 Back's kight, fdg his courage faggg, begs to whiste ervousy ad head for the kgside, where his oppoet makes troube.

55 g5 fxg5 56 xg5 The pa is ucovered, strata by strata. 56  Nf3 The oss of g5 is irreevat.

57 Kb4 Nxg5 58 e5 The crowd sways, mutters ad swes. The metamorphosis s amost compete ad soo the crowd surges forward  its ew, terrifyg form: a ych mob. 58  h5 59 e6+ White's kight chaperoes the d-ad e-paw pair up the board, as guards woud escort upredictaby dagerous mates from their ce bock to a ew maximum security ocatio. 59  K8 60 Kxb5 1-0 Game 55 MBotvk-DBste Word Champioship (7th matchgam), Moscow 1951 Dutch Dence 1 4 5 2 c4 e6 3 Nf3 c6 4 g3 f5 5 Bg2 N 6 0-0 Be7 7 b3 0-0 8 Ba3 b6 9 Bxe7 Qxe7 10 Ne5 Bb7 11 N2 Nb7 12 Nx7 Nx7 13 e3 Rac8 14 Rc1 c5 15 Qe2 N 16 cx5 Bx5 17 Bx5 ex5 18 Nf3 Rc7 19 Rc2 Rfc8 20 Rfc1 Ne4 21 Ne5 Nf6 22 Q3 g6 23 Qa6 Kg7 24 Qe2 Q6 25 a4 Ne8 26 Q2 N 27 Qc3 Ne4 28 Q3 cx4 29 ex4 a5 30 Kg2 N 31 Qe2 f4 32 gxf4 Nh5 33 Rxc7+ Rxc7 3 Rxc7+ Qxc7 35 Qg4 N 36 Qe6 Nh5 37 Q7+ Qx7 38 Nx7 Nxf4+





Botvik eters a kight edg, about to be a paw up.

39 Kf3 Nd3 40 Nxb6 Nb4 Back's a-ad d-paws have bee secured for ow. The most terrifyg words from a prospective pubisher to a writer are these: "Whie your writg shows great promse, at this time we fd that it fais to meet our commerciay viabe threshod

for pubicatio. These sad otices remd me a bit of Brostes positio, which comes tatagy cose to draws, but misses  each key e. He does his best  this edgame, but there s o path to save himse agast accurate hite pay.

41 Kf4 Kf6

 (  ): Whit is a paw up ad hods a dg with kg

positio as we. However,

hs two to oe queeside paw majority has bee frmy bockaded. How woud you make progress here?

 Back's dot kight must b chagd via 5 d d3.

42 Nd7+! Ke7 43 Ne5 Ke6 44 Ke3! Rarey do we see retrograde kg stro  a edg, but here ejectio of Back's kight represets the top priority. 44  Kf5 45 f3!

: What is th pot of Whit's ast mov?

 Whit bocks off . . . Kg4 fitratio attmpts wh h movs his kight to d3.

45  g5 : Why ot 45 ... Kg5, aftr which Whit's

kg must stay put ad guard the

kgside?

 Th ida gags  a fas suppositio that Back hods his ow  a promotio rac. Botvk had workd out th w for Whit o yor suggstio: 46 Kd2! (trg th gam to a hair-raisg scramb) 46 . . . 4 (aftr 46 ... Kf4? 47 Nd3+ Nxd3 48 Kxd3 Kxf3 49 b4, Whit is much fastr  th kg ad paw dg) 47 Nd3 Na6 48 Kc3! 3 49 b4 Kxh2 50 bxa5 h5 51 Nb4 Nc7 (or 51 . . . Nb8 52 Nxd5! Kg3 53 N3 Kxf3 54 d5 ! !, wh Back's kight provs qua to th task of hatg th srgg a-ad d-paws) 52 Nxd5 Nxd5+ 53 Kc4 Kg3 (or 53 ... N7 54 a6 Nc8 55 Kc5 h4 56 d5 Kg 57 d6 Nxd6 58 a7! ad ws) 54 Kxd5 h4 55 a6 h3 56 a7 h2 57 a8Q hQ 58 K6 with a wg qu dgam for Whit.

46 Kd2!

46  hS Oce agai, Back oses the race after 46 . . . Kf4?? 47 Nd3+ ! (the kight's eyes arrow as they trace the route of his teded victims  he reaes d3 s perfect terra for a ambush) 47 . . . Nxd3 48 Kxd3 Kxf3 49 b4.

47 Nd3 Na6 Or 47 . . . Nc6 48 Kc3 foowed by b3-b4.

48 Nc5 Nb4 49 Nd3 The trademark Botvk repetitio of moves. 49  Na6 50 h3! Back threateed . . . g5-g4. 50  Nc7 The kight receives trifg remueratio for og hours of toi. 51 Ke3! "My mistake, says hite's kg  hs vaium voice. His majesty cutivates his busess with a ar of casua egect, whe  reaity he bides his time, awaitg the proper momet to strike. Botvik disguises  true tet with a few head-fakes o the kgside, whie by tossg  a few radom kg moves, he draws coser to the time cotro without aterg the positio. 51  Na6 52 Ke2





Back's kg must give groud sce his kight must cotue to rema vigiat agast b3-b4. 52 ... Ke6 53 Kd2! Ke7! Back's kg pus a about-face ad heads  what appears to be the wrog drectio. Broste sees through Botvk's tricks, as if with x-ray visio.

: To rtrat sms crazy. Why ot mov up with 53 . . . Kf5 - ?

 It fais to hp Back ftr 54 Kc3 Kf6 (54 ... K6 oss staty to 55 Nc5+, wh Whit's kight pus out his ridg whip  to th costratio of Back's kg ad kight, thy simutaousy raz that Whit's kight dos't ow a hors!) 55 b4 axb4+ 56 Nxb4 Nc7 57 a Kf5 58 Kd3 Kf4 59 K2 Kg3 (th kg tosss moy ad rsorcs o this jt to owhr; stad 59 ... Kf5 60 a6 K6 61 a7 rachs a situatio siiar to what happd  th gam   a way this variatio capsuats Brost's ack of good fort  th tr dg) 60 a6 Kxh3 61 a7 (th schoo buy o a7 grabs th adjact, dfdg kight, trs him upsid dow by th aks d shaks him,  ordr to mpty his pockts of ch moy) 61 ... Kg2 62 Nxd5! Na8 (th bat dow kight, tru to his rigious bifs of o-vioc, trs th stadium, ad ss saivatg ios ad chrg haths) 63 N3+ Kg3 64 d5 ad ws.

54 Kc3





At ast, White's kg heads for the aggrieved fak, ready for b3-b4, without fear

of Back's kg fitratg the kgside. 54  Kd6 55 b4 After refectio ad cautious appraisa, Botvik foows through with hs origa pa to activate his queeside majority. 55   axb4+ 56 Nxb4 Nc7 57 as Nb5+ 58 Kd3 Ke6 59 Ke3 The white kg comes at Back  two drectios,  scssor formatio. 59  Na7 The retreatg kight waks with the eade steps of the eight-year-od takg that og, sow pigrimage to the prcipa's office. 60 a6 Nb5 61 Nc6! Seg cotro over a7. 61  Nc7 Isomiacs are ofte kept awake by ther ow fear of ot fag aseep. Back's kight must rema ever vigiat to the passed a-paws threat to promote .

62 Nb4 More time gaed o the cock. 62  Kf5 63 a7 Ke6 Back's kg is the office worker steg to his hated boss' ufuy joke, patiety awaitg its ed, so that he ca fake a hearty augh ad put o a scere, amused expressio. 64 Kf2! Itedg Kg3, foowed by h3-h4! which creates a secod passed paw. 64  h4

The reetess desire to aiate a wouded eemy is the basest ad most primordia of emotios. Back's kight, a cassic exampe of defesive caustrophobia, ow boud ad gagged, must rema  exie to cover agast White's queeg threat.

 (  ): I ordr to w, hit's kg must

participate  the fight. How to ga

etry to Back's camp?

A Caracovroad. Tmporary sac a paw to crat a avu to th kgsid.

65 f4! gxf4 66 Kf3 1-0 c 66 ... 5 is mt by 67 Nxd5!, ovroadg Back's kight.

Game 5 SReshevsky-MBotvk  vs. R atch, Moscow 1946 rench Defence 1 e4 e6 2 4 5 3 Nc3 Bb4 4 e5 c5 5 a3 Ba5 6 Qg4 Ne7 7 xc5 Bxc3+ 8 bxc3 N7 9 Qxg7 Rg8 10 Qxh7 Nxe5 11 Be2 Qa5 12 B2 Qxc5 13 Nf3 Nxf3+ 14 Bxf3 e5 15 Bh5 Bf5 16 Bxf7+ K7 17 Qh6 Rxg2 18 R Qb6 19 Qxb6 axb6

: It ooks to m k Whit is i troub

from the start of ths edig. Woud you

agree?

1. White s a paw up but this IS a purey temporary state, sice Back attacks both c2 ad h2. 2. White ows the bishop pair i a competey ope positio. 3. White's paws across the board are astersked with mutipe structura defects. 4. Back's kg s more active, as are his rooks. 5. Back cotros the cetre but may be vuerabe ater o to f2-f4 or c3-c4 breaks. Cocusio: Despite appearaces, the positio ooks coser to eve. 20 0-0-0!  This is a tough o to assss:

: Why did Whit giv away his vauab,

passed h-paw whe he coud simpy

push it?

 Lt's ook: 20 h h8 21 Bg5 h7! (Whit's ight-squard bishop has o pac to go) 22 Bx7 xf7

23 Bg5 Bxc2 ad ow t's assss:

1. Materia is eve but Back's forces are ceary more

active tha White's. 2. The presece of opposite-cooured bshops may ot actuay hep White here, sce his ight squares ook weak ad vuerabe to fitratio. Cocusio: Your e is payabe, but it s White who fights for the draw at the ed of it. 20  Rxa3 20 . . . Rxh2 21 Bg5! Rxa3 22 Kb2 Ra5 23 Bxe7 Kxe7 24 Bxd5 ooks ike a draw.

21 Kb2 Ra4 22 Be3 Doube attack o d5 ad b6. 22  Be6

23 Bxe6+ Vioatg the prcipe: Don't swap one of your bishops o  you already own the bishop  par

: o why did Rshvsky tr this ?

 H probaby did't trust 23 Bh5 xh2 24 B2 Kc6, wh Back has a xtra paw. Idd, I'm ot so sur Whit's bishop pair provids fu compsatio.

23   Kxe6 24 Bxb6 Rxh2 25 Rgl The positio  a ikey draw with so few paws remag o the board. 25  Rh6 I order to cover g6 ad free his e7-kight.

26 Rg7 Rg6 27 R7 Now Back's b-paw   dager sce Bd8 is  the ar. 27   Nf5!? A typica Botvk decisio, offerg a paw to activate his kight ad trasfer it to the domatg c4-square, abeit a pa desiged to timidate more tha fict actua damage. The Joker character from the Batma movies was ever reay motivated by thoughts of moetary ga. He simpy wated to spread terror ad bow thigs up.

28 Rxb7 Nd6 29 Kb3 29 Rb8 Nc4+ 30 Kb3 Ra3+ 31 Kb4 Ra2 32 Kc5 Rxc2 33 Rxd5 Rxc3 34 Rdd8 Nd6+ 35 Kb4 Rc6 shoud ed i a draw. 29   Ra8 30 Rc7 Rb8 31 Rc6 Kd7 32 Rc7+ Ke6 33 Rc6 Rb7!? Botvik pays for the w a paw dow. 34 c4! Reshevsky wsey offers a paw  retur to get his kg out of the box. 34  dxc4+ 35 Kb4 Ke7 36 Ka5!? Reshevsky pays for the w as we. Now Bc5 is meaced. Perhaps the issue of payg for the fu pot is oe which both parties ca agree upo, eve if they cat agree upo aythig ee. 36 Rc7+ Rxc7 37 Bxc7 Nf7 s a ear certa draw. 36   Kd7 37 Rxc4 Re6 38 Ka6 Rb8 39 Rc7+ Ke8 40 Ka7! Rd8 41 Rh1?? 41 Rb1 Rd7 shoud be draw despite ite's extra paw.





After some briiat pay, Reshevsky, perhaps ot reaizg he had aready made the time cotro, bags out a extra move which turs out to be a terribe buder.

 (  ): This o is ot as asy as

fork o either b5 or o c8. Oy oe of them s correct.

it ooks! Back ca ow

41  Nb5+? Fork. But ot ot o the right square! Botvik edagers his wi with a retued accuracy.  Back ws with 41 ... NcS+! (th corrct squar) 42 xcS (42 Ka6 Rxb6+ 43 Ka5 fas to 43 ... Rd5+ 44 Ka4 KdS! 45 f7 N7, scapg th atg t) 42 . . . xcS 43 hS+ Kd7 44 xcS KxcS (th pot: Whits kg is bady cut off, th rsut of which is h drops aothr paw by forc) 45 c4 (45 Ka6 Rc6 is quay hopss) 45 ... Rc6! 46 c f6! ad Whit is bustd.

42 Kb7 Nxc7

Reshevsky, by ow competey fustered at mssg a simpe fork, budered aga. The aperture of drawg opportuity coses rapidy upo White ad we reach a divide. We aways beieve we have choices, but sometimes there s just oe way ad a others ose, therefore dsquaified as true choices.

 ( ): houd Whit pay th mdiat rcaptur with

43 Bxc7 - ? Or shoud

he toss  a rook check first with 43 Rh8+ - ? Choose carefuy.

43 Bxc7? The wrog way. By aowg both back rooks to rema o the board, Reshevsky uderestimates the dagers to his kg.  H shoud hav go for 43 hS+! Kd7 44 h7+ R7 45 x7+ Kx7 46 Bxc7 Rd7 47 Kc6 4 4S Bf4 K6 49 B3 ad Whit shoud draw without ay troub.

43  Rd4! 44 c3 Rc4 It st obvious at this pot but White's kg remas  dager. Botvik breathes ife to  edgame matg attack, the way a scuptor imparts ife to raw stoe through chise, hammer ad imagatio. 45 Ba5 Kd7 46 Rh8 Rf6 47 Rd8+ Ke7 48 Rd2 Rd6! 49 Ra2 Now the trade of rooks comes too ate: 49 Rxd6? (49 Bb4?? of course rus to 49 ... Rxb4+ 50 cxb4 Rxd2) 49 ... Kxd6 50 Kb6 Kd5 51 Kb5 Rc8 52 Bb4 Rb8+ 53 Ka5 Rb7! ad there is o good way to stop ... Rf7, pickg up the f-paw. 49   Kd7! 50 Rb2

·



Despite the u, a of Reshevsky's pieces have fae competey out of syc, whie every back piece harasses White's kg. Botvik forced the w of a paw by combg it with matg threats, with a probem-ike move. Back's remag forces brste, the way my haughty dogs behave whe our mai carrier deivers a package.

 ( ): W must discovr th hidd ocus of Whit's soft

spot  the

positio. Ca you fd the geometric aomay Botvk foud?  50   R! ! 51 Bb6

"God s testg me, the bishop mutters to himse. For this impovershed mok, ife s perpetua Let. There is a differece betwee a deiberate, reorgaizatioa retreat ad a fu bow rout, but  ths stace the differece becomes burred. White's positio, by ow permeated with fuerea goom, gets eve more depressg. White had o choice ayway: a) 51 Bb4?? (the bshop caot eave c7 uguarded) 51 ... Rc7+ 52 Ka8 Ra6+ 53 Kb8 Rb6+ 54 Ka8 Kc8! forces mate. b) 51 Ra2?? Rb5+ 52 Ka7 (the kg fais about  reewed agitatio  to a fugitive trapped ad surrouded, food, water ad ammuitio s as ecessary as breath itse) 52 ... Kc8! 53 Bb4 Rb7+ 54 Ka8 Rb8+ 55 Ka7 Rd7+ 56 Ka6 (Death's messeger hovers ear White's kg, who coses his eyes, as if the actio makes his assaiats dsappear) 56 ... Rdb7!, whe there s o defece to the comg ... Ra8+ mate. 51 000 Rxc3 52 Rb4 Ke6 53 Rb2 Rdd3! Itedg . . . Rb3. Now Back wats rooks off the board. 54 Ra2 Rd7+ 54 . . . Rb3 is ao wg.

55 Ka6 Rb3 Threateg . . . Rd6. 56 Be3 Rd6+ 57 Ka5 Rd8! Threateg ... Ra8+.

58 Ka6 If 58 Ka4 Rb7 ws, whie 58 Ba7 Ra8 59 Ka6 Kf5 60 Ra4 Rb2 61 Ra5 Rxf2 62 Kb7 Rxa7+ 63 Rxa7 Rc2 s a easy w for Back as we sce White's kg is AWOL.

 ( ): Whit pas to bock a rook chck o a

with Ba7, whe the bshop, a frighteed chipmuk, races dow the decivity to its hidde est. hat did ite mss  this positio?

 Rmova of a dfdr. Th gia, avcuar ighbor o b3 trs out to b th sria kir th poic hav b sarchg for.

58 000 Rxe3! 0-1 No mor bocks o a7. Whit's bishop ss his bssgs at a high pric, but th rook is wg to pay. Game 57 MBok-MTal Word Champioship (11th matchgam), Moscow 1961 lav Defence           

: Why a xcam for th du Exchag av?

 cisy bcaus of its hrt duss. Who is bttr suitd to th rsutg static, tchica positios arisg? Botvk th mostr tchicia, or Ta th tactica wzard? I thk this was a poor opg choic o Tas (ad his coach's) part.

4 000 cxd5 5 Nf3 Nc6 6 Bf4 Bf5 7 e3 e6 8 Bb5 Bb4 9 Ne5 Qa5 10 Bxc6+ bxc6 11 0-0 Bxc3 12 bxc3 Qxc3 13 Qcl Qxcl 14 Rfxcl

:  dad draw?

 Not qit so dad, dspit th prsc of opposit-coord bishops. Whit hods th advatag sc h ads  dvopmt, ad his kight, oc it rsts o c6 w b vry dffict for Back to jct.

14  0-0 15 f3!

: Why did't h captr th c-paw?

 Th c-paw is't gog aywhr, so Botvk pays his ast mov with for ttios:

1. He

prepares to activate  kig via f2. 2. He deies Back use of the e4-square. 3. He eaves the possibiity ope for the space-gag g2-g4. 4. Lasty, he tempts Back to ... Nh, which favours ite. 15  h6 : What is so wrog with 15 . . . Nh - ?

 It bgs a frior dg aftr 16 Nxc6 Nxf4 17 xf4 wh Whit's kight doats o its c6 post. Back is  dagr of droppg his a-paw atr to a Rc3-a3 mouvr.

Nor does Back equaze after 15 ... Rfc8 16 g4 Bg6 17 h4 h 18 g Ne8 19 Nxg6 fxg6 20 Rabl, whe hite threates Rb7 ad Be, with a parayzg bd, despite hs oe paw deficit. 16 Nxc6 Rfe8 17 a4! Perhaps tedg a4-a, cotempatg the hat of . . . Nd7-b6 ideas . 17  Nd7 18 Bd6! Eve stroger tha the thematic 18 a, sce Botvk's move sets up a trap, which Ta waks to. 18  Nb6?!

 (  ): How did Botvk improv his positio?

 Trasfr th bishop to c d thrat Nxa7! .

19 BcS! Bd3 Ateratives: a) 19 ... Nc4 20 e4! Bg6 21 exd exd 22 Ne7+ Kh7 23 Nxd picks off a importat paw. b) 19 ... Nc8 20 Ra3! (cuttig off ... Bd3, whie preparig to trasfer his rook uchaeged to the b-fie) 20 ... a 21 Rb3 ad Back ca barey move. 20 Nxa7! Hdini prfrs th facir  20 N Bh7 21 Bxb6! axb6 22 Nd7 Ra6 23 Rc6 Ra8 24 Nxb6, wh Whit avoids drawg pasatss basd upo bishops of opposit coors. O th dowsid, Whit's forcs ook a bit tagd up.

20 000 Rxa7 21 Bxb6 Ra6 22 as Bc4

: I drstad that Whit is up a paw, but Back hods th trump of th opposit-coord bishops ad a compty bockadd a-paw. Ca Whit w?

1. There may be the presece of opposite-cooured bishops, but White sti stads a cea paw up. 2. The passed a-paw, athough bockaded as you metioed, has the beeficia effect of tyig dow two of Back's pieces. 3. White ca eve try a pa of oadig up rooks o the c-fie ad perhaps trasferrig the kig to b4. The he may pu off a exchage sac o Back's bshop o c4, which woud eave him with two paws for the exchage ad at east oe back rook competey tied dow to the a-paw. Cocusio: The presece of opposite-cooured bishops does itte to subside Back's apprehesios, ad i o way assures him of hodig a draw.  It wot b asy but Botvnk has puss as w:

23 Ra3 £6 24 e4 Botvik begs to gai kigside space, sce Back ca do othig but await evets. 24 000 Kf7 25 Kf2 Raa8 26 Ke3 Re b8 Threateg . . . Rxb6. 27 Rac3! The exchage sac o c4, if ot ow, the at some poit i the future, becomes a serious issue for Ta, who decides to ivest the other rook o c8 to prevet it. 27 000 Rc8 28 g4 Rab8 29 h4 Rc6 30 h5 Botvik cotiues to aex territory. 30 000 Rbc8 31 e5 g6?





A pa which overpromises ad uder deivers. Oh, the sufferig of watg (couterpay) ad ot havg it. Ta reads the positio  the ight of beigerece, hopg to ray with the reewed ferocity of oe who aready cosiders himse amog the dead. Yet deep with him a voice arses: "If ife remas, the so does hope. Ta's diemma is that of a kg who reaizes there may be riots if he decares war, ad possibe riots if he doest! Passivey awaitg evets was ceary ot Ta's strog suit. Hs md s made up, ow impervious to sway from foy. So he duges  a preposterous risk, with oy dimishig chaces of retur. The troube that is ay disturbace of the kgside paws by Back oy heps Botvik's side. I this case, Ta's move criticay weakes h6, after which hope dsappears, to be repaced with despair. Soo Back's positio sags ad the dark imputatio fas upo hs dubious 31st move, the atura suspect  the crime.

32 hxg6+ Kxg6 33 R3c2! Targetig h6 after ite doubes rooks aog the h-fie, which offers ready access ito the eemy camp. The coaescece of Back's strategic woes grows to decisive proportios . 33  fxe5 : Ta jst hadd Botvik a kgsid

paw majority. hy ot just shuffe ad

wait?

 t this stag thr was othg bttr. For xamp: 33 ... Ra8 34 xf6 Kxf6 35 h1 Kg6 36 Rch2 (th rooks p up o h6 k a co of hgry ts  sarch of food) 36 ... h8 37 h5 Rcc8 38 Kf4 Rcf8+ 39 Kg3 Bd3 40 Bc7 Ra8 41 g5, wg a scod paw.

34 dxe5 Rh8 35 Rh2 Rcc8 36 Kd2! The kig decides to accommodate by steppig out of the way, ad cearig e3 for hs bishop to pie up o h6. 36  Bb3?





A sete is't much of a sete  he rus at the frst sig of troube. Ta gives vet to resetmet with fruitess aggressio. I his edeavour to achieve activity, he aows Botvk's a-paw to dig yet deeper. It becomes pafuy cear that Back's "itiative is fashioed from buster ad facade, devovg with each move.

37 a6 Botvik ist about to fa for 37 Rxc8 Rxc8 38 a6?? Rc2+, pickg off hite's rook. 37  Bc4 38 a7 This paw has bee drive  too deep ad Back must keep a costat eye out for promotio schemes, as we as vest pieces to hat the paws further ambitios. 38  Rh7 39 Ral Ra8 40 Be3

·



The ship's gu barres sowy pivot right, soo to dscharge the fata cargo at h6, where Back's pieces have bee drive ito pockets of utter hepessess. Note the bshop's dua drectioa efficiecy: it attacks h6 whie protectg the a7-paw. 40  Rb7 40 . . . Bb5 waks to a overoad combatio after 41 Rah1 Rah8 42 Rh6+ ! . 41 Rxh6+ Kg7 42 Rah1 Rb2+ 1-0 Back's kg is swpt to th vortx of hit's attack ad 43 Kc1 Ra2  7+ ds it. Game 58 YAverbakh-M Botvk Trag match, Moscow 1956 icilian Dence 1 e4 c5 2 Nf3 Nc6 3 4 cx4 4 Nx4 N 5 Nc3 6 6 Bc4 e6 7 a3 Be7 8 0-0 0-0 9 Ba2 B7 10 Be3 Nx4 11 Bx4 b5 12 Q3 a5 13 Nxb5 Nxe4 14 a4 Bc6 15 Ra1 5 16 c4 xc4 17 Bxc4 Qb8 18 Qf3 Qb7 19 Rfe1 Rf8 20 B3 Nf6 21 Qh3 h6 22 Bc4 R7 23 Be5 Ra8 24 Rx7 Rx7 25 Qg3 B5 26 Be2 Ne8 27 Rc1 Bg5 28 f4 Bf6 29 B B8 30 h3 Be4 31 Kh2 h5 32 Qc3 Bb6 33  Qa8 34 bxa5 Bxa5 35 Qc8 R8 36 Qxa8 Rxa8

: It ooks to m k hit is th oy o with

paw, correct?

wg chaces due to hs passed a

 It ooks that way to m too. I fact, it ooks k Whit c v drop his a-paw ad st mayb maag a draw. Yt Botv wo agast a strog gradmastr (ad a dgam spciaist to boot), may du to Back's hidd pus: Whit fds it dfficut to covr his scod rk.  th sam, with carfu pay, Whit shoud st hod a dg du to th powr of his passd a-paw.

37 g3 Rd8! Threateg to vade the seveth rak.

38 Bc3 Bxc3 39 Rxc3 Logica ad probaby accurate: White covers g3, the softest spot   positio ad thiks about swgg hs rook to a3. White's pieces beg to drift away from their goa, as  respodg to a summos oy they ca hear. After the correct 39 Nxc3! Rd2+ 40 Kg1 Bf3 41 a Nd6 42 a6!, my feeg is that White is the oy oe with wg chaces, sce if Back ssts o 42 . . . Nf White mee ts it with 43 Nb! ad the g-paw cat be touched due to Rc3. 39  Rd2+ 40 Kg1 Kf8 41 as Bb7

42 Na3?! Last chace for 42 Nc7! Nd6 43 a6 Be4 44 Nb! which reaches a positio simiar to the (favourabe for White) previous ote . 42  Rd1 Botvik prods here, pokes there, as if to awake a dormat idea he does't have ust yet. 43 Rb3 Ba6 44 Rb1 Rd5 45 Nb5 White fusses over his a-paw, the way my eighbour washes his spotessy cea car ad the ovgy swabs it dow, dryg it with ovg caresses of a ShamWow towe. 45  g5! Botvik stregthes his positio, activatg hs kgside paw majority. Back is o oger worse.

46 fxg5 Rxg5 47 Kf2 The kg reuctaty prepares to take eave of his home with a sese of meachoy. 47  Rf5+ 48 Ke3? Uswervg, resoute ad correct! White shoud have goe back aga with 48 Kg1.

·



The ervous ab rat approaches, reaizg that if he steps upo the correct ever, he gets rewarded with a chuk of cheese;  he steps o the wrog oe, a eectric shock. For a atura optmist, the word seems reewed ad a thgs possibe. White attempts to pay his debts with bue, $50 Moopoy bis, which ted to be frowed upo as ega currecy outside the Moopoy word. His ast move foowed the prcipe: Centralize your kin g in an endin g. However, chess is a ficke game ad it turs out this positio is the exceptio to the orm.

 ( ): Th partis striv for cotro

over b with mutua opportuism,

but  ths case Botvk expoited the geometry to pick off White's dagerous a-paw. How?

 tp 1: Chop th kight.

48  Bxb5! 49 Bxb5 This is the troube with 48 Ke3?  White ca't recapture with the rook. At this pot, Averbakh may have reaized he missed Botvk's comg trick  his cacuatios. Step 2: Attack the bishop ad pick up a. 49  Nd6 The defece is breached, ike a trash bag which ow eaks ad oozes its ve, foody residue of goo, cosistg of smey tua, ketchup ad rotte cabbage. 50 Bd3 Re5+ 51 Kd4 The kg owers his head  cotritio, promisg to chage hs ways, but the syy gives a cospratoria wk ad puges forward yet aga. He siffs cotemptuousy at the out o e, approachg hm with pushmet  md. Oce aga this proves to be the exceptio to the prcipe.

: Why is it corrct to ctraiz hr?

 Whit's kg is dd at hom to dfd his paws. Whit was bttr off payg 51 2.

51   Rxa5 52 g4? Questio: Why a questio mark? White foows the edgame prcipe: The defendin g side should redu the number of pawns on the board.

  thrd prcip foowd ad a third xcptio to th ru. How frustratg for Whit  h dutfuy foows protoco ad fds himsf dr cotua rprmad. Th raso: tactics aways tak prcdt ovr prcips. fortaty for Whit,  this stc h oy aags to rduc his ow paws, ot Back's! Th mov is a bdr which drops a paw. H shoud backtrack aga with 52 K3.

52 ... hxg4 53 hxg4





We sese movemet from the shadows. Botvik has a fia, mordat itte joke with a asty twst at its competio.

 ( ): Back to pay ad w aothr paw.

  skid htr oy rquirs partia sight of his pry to tak th shot.

53 ... Ra4+ 54 Kc5 "What s the meag of ths! beows the kig, ametig the fact that word evets do ot traspire to  ikig, despite his great efforts to impose hs wi upo them. He is o high aert, the way a Huro scout spots Iroquois moccas imprts o the trai. 54 ... Rxg4! 0-1 Th pot: 55 Kxd6 is mt by 55 ... Rd4+, rgag th pic with trst. Ivstors from Whit's sid ract to th ws of sas rg fat ad quartry rports bar aarg tidgs of rcord dptios of stock vau. Game 59 MBotvk-DBste Word Champioship (23rd matchgam), Moscow 1951 G nfeld Defence 1 4 Nf6 2 c4 g6 3 g3 c6 4 Bg2 5 5 cx5 cx5 6 Nc3 Bg7 7 Nh3 Bxh3 8 Bxh3 Nc6 9 Bg2 e6 10 e3 0-0 11 B2 Rc8 12 0-0 N7 13 Ne2 Qb6 14 Bc3 Rf8 15 Nf4 Nf6 16 Qb3 Ne4 17 Qxb6 axb6

·



Now begs oe of the most dramatic edgs  word champioship history. Botvik was a pot dow with oy this game ad the ext remag. He had to score 1 out of 2, to tie the match ad reta his tite  ad this was his fa shot with the white pieces. So, with such tese pressure o him to w, Botvik paced hs hopes o his edgame skis.

18 Bel : Why rtrat a bad bishop to avoid th swap?

 Bad bishop or ot, Whit must rta som kd of mbaac  this positio i h is to hav ay hop of victory. Th structur for ow ay b rigid, favourg Back's kights, but th gam is og ad thgs chag. If th gam ops up atr o, Whit's bishops may com to f.

18  Na5 Itroducg the . . . Rc2 idea. GM Ludek Pachma writes: "I we remember that most commetators had a poor opio of Botvk's positio at this pot.

: Dos Whit stad wors?

 I do't thk so sc, as mtiod, Whit's bishops, athough rathr sorry-ookg for ow, may atr sh i th gam ops. Th positio is probaby dyaicay baacd at this stag.

19 Nd3 Bf8 : Why did't Brost go for th mmdiat vasio of c2?

 Th imdiat 19 ... Rc2 c b mt by 20 Nb4! d i Back sists with 20 ... xb2?, h drops a xchag aftr 21 Bx4! dx4 22 Bc3 R2 23 d RcS 24 Kf1 xc3 25 Kx2.

Botvik ao caimed a edge for ite at the ed of the e 19 ... Nc4! 20 Bb4 Bf8 21 Bxf8 Kxf8 22 Rfcl, but to my eyes, ths oe ooks very difficut to try ad w from ite's side.

20 f3 N d6 21 Bf2 B6

21 . . . Rc2 is met by 22 Rfcl Rdc8 23 Rxc2 Rxc2 24 Rcl, whe the vasio party s repeed.

22 Racl N ac4 23 Rfel

 (  ): It is  Back's bst trst to forc Whit

to a rigid paw structure.

How ca he accompish ths feat?

23  Na5?! Broste aows himse to be dstracted from hs madate by markg time, perhaps thikg aythg draws.  I hav a fg h woud hav wo th word champioship i h had payd th simp 23 ... Nf!, crasg th prssr upo 3, ad forcg th pasat 24 f4, aftr which Whit's wg chacs go up  a haz of smok. Back ay actuay stad bttr aftr 24 ... Nfd6, ad h crtay dos't ook k h ca os hr.

24 Kfl Cetraizg hs kg. 24  Bg7 25 g4! Cuttg off . . . Nf5 ideas. 25  Nc6 26 b3?! Botvik rightfuy criticized his move, which uecessariy weakes both a3 ad c3. 26  Nb5! Broste immediatey targets the weakeed squares. 27 Ke2?! 27 a4 shoud be tossed . 27  Bf?! Broste wats too much from the positio, hopg to have time for ... Ba3.

: What shoud h pay stad?

 H had 27 ... Na3! foowd by ... BfS!, sg th qusid dark squars, aftr which Back may aga v stad bttr, but qua at th vry mimum.

28 a4!





Rectfyg his accuracy o his previous move. 28  Nc7! I order to trasfer to a6 ad the possiby b4.

29 Bg3 The oce-sow bshop begs to she, ke the schooboy who the teacher beieved was the duce of the cass, who ow shocks her with ewy dscovered aptitude . 29  Na6 30 B£l £6 31 Redl Na5 Broste cotuay probes the soft pots  hite's positio. Botvik s pretty much forced to a paw sac, for which he receives ampe compesatio. 32 Rxc8 Rxc8 33 Rcl Rxcl 34 Nxcl Both sides had foresee the comg positio ad assessed it differety. Back ca force the w of a paw by attackg the sge defeder of b3. 34  Ba3 35 Kdl

 ( ): Woud you chop o c to w

Back's bishop?

the paw, or avoid it ad hag o to

35  Bxcl?! Nacy Reaga woud agree: "Just say o! Here Broste had to do somethg extraordary  othg! His move ooks too mercatie  toe for the actua requiremets of the positio. I have a feeg this move is where Broste aowed the word tite to sip away. White probaby cat w if Back eaves the positio as is, recogg that further aggressive actio proves uavaig to his greater cause of hodg the draw. 36 Kxcl Nxb3+ 37 Kc2 Na5





White's petifu compesatio for the paw: 1. The bshop pair  a rapidy opeg positio. 2. A fuid paw structure, which favours his bishops. 3. A sighty more active kg. 4. The a6-kight, through o faut of his ow, fds himse impicated  hs bshop's crime. It doest have a sge move after White's ext, appearg reuctat, ike a ace-faced teeager, s e-coscious about posg for the schoo photo. 5. Back's extra paw is that ame, soated, doubed guy o b6. It comforts yet fai to hep, a weak fa o a hot summer ight, merey heard but ot fet  which meas that, for ow, White's a4-paw hods back both back b-paws. So  essece, Back gave up a powerfu dark-squared bshop to w a worthess paw. Cocusio: After hs dubious decsio to take the b3-paw, Back must ow fight for the draw. 38 Kc3! Hemmg  the a6-kight. 38  Kf7 39 e4 f5?! Back's paws form a ugy arch, a od ha-eate doughut, tossed  a dumpster. Broste wated to force the ssue ad carify the paw structure, but this move oy maages to crease the power of White's bshops by opeg the positio.

40 f5 f5 41 Bd3 Kg6

·



Aother game with two critica decsio exercises. Image the pressure: Botvik must w this game for a reaistic shot to reta his word champioship tite. He frustratgy seses the wg pa urkg, yet expressibe ad uidetifiabe  eary o-existet, yet papaby there. Botvk attempts to grasp somethig from othg, the way a ward creates from pure empty space with a mere catatio. White's itiative fuctios ear capacity. The questio is how ca we tweak it to get ust a itte more kick?

 ( ): Whit ca pay 42 Bb, gog aftr d5, or h

ca try to crease the

pressure with 42 Bd6, mprisog the a6-kight. If you were  Botvik's shoes, which move woud you pay?

42 Bd6? The seaed move ad a correct oe. The coceivabe pressure gets to Botvik, as he picks a correct path. A discouraged Botvik, whe eavg the payg stage, was heard to te a member of hs aaytica team: "A s ost ... !  Whit has vry chac to w th gam aftr 42 Bb! Nc6 (ot 42 ... Nc4? 43 xf5+ xf5 44 Bf4!, thratg Ba2 ad ws) 43 xf5+ xf5 44 Ba2 (th bishop shows hmsf from his pac of hidg, bgg to prforat th dfsiv wa via th xtrmitis) 44 ... N7 (ot 44 ... Nab4? 45 Bb3 with Bd6 xt) 45 Bh4 Kf7 46 Bx7 Kx7 47 Bxd5 ad Whit picks off b7, with vry chac to covrt to victory.

42  Nc6 43 Bbl





The bshop takes his eave. Back soo ears a bitter esso: just because a eemy s goe doest mea he has dsappeared. 43  Kf6? Botvik himsef poited out the correct drawig pathway: the semi- m ea lpa cotortio 43 ... Na7!! 44 exd5 exd5 45 Ba2 b5! 46 a b4+! 47 Kd3 Nb5 (the kight emerges a free ma with  priso setece remitted) 48 Be5 Nac7 ad Back shoud hod the game. 44 Bg3! Botvik retas sef-possessio eve uder great duress, producig the ripeed fruit of adjourmet aaysis with the trickiest move  the positio. Istead, 44 Bf4? ! Ne7 shoud draw. 44  fxe4? Actuatg a cha of evets which fow forth by impuse. Broste vioates the prcipe: Don't open the position when the opponent has the bishop pair Kasparov caims Back shoud sti draw  the e 44 . . . h6! 45 Bf4 h5 46 exd5 exd5 47 h4 Ke6 . 45 fxe4 h6 46 Bf4 h5 47 exd5 exd5 48 h4!





Dua purpose : Aowg the bishop etry to g5 ad fixg h5 as a statioary target. 48  Nab8 After 48 ... Ne7 49 Bg5+ Ke6 50 Bxe7 Kxe7 51 Bg6, White shoud w due to the cumuative effect of: 1. White's soo to be passe d h-paw. 2. The bshop's superiority over the kight.

3. Back sti has target paws o d5 ad c8. 49 Bg5+ Kf7 50 Bf5! The bishop's right ip cur ito a hatefu haf-smie at the thought of deyg Back . . . Nd7. 50   Na7 51 Bf4 Nbc6 52 Bd3! Dua purpose : Prevetg . . . b6-b5 ad preparg to reoad o h5, via e2. 52  Nc8 Or 52 ... Ne7 53 Be2! Kg6 54 Bg5 Nac6 55 Bf3! (zugzwag!) 55 ... Nf5 56 Bxd5 Nfxd4 57 Be3 ! Nf5 58 Bxb6 shoud be eough for ite to w, sce 58 . . . Nxh4?? is met by 59 Bf2! Nf5 60 Bxc6 bxc6 61 a Nd6.

-

-

Visuay, this ooks ike it shoud be a easy draw for Back, but due to a quirk  the geometry, Back's kight simpy ever gets a chaces to sac itsef for ite's oe a-paw: 62 a6 Kf5 (62 ... c 63 Bg3 Nc8 64 Kc4 Kf5 65 Kxc5 Ke6 66 Kc6 s simiar to the ma variatio, whie 65 . . . Ke4 66 Kc6 Kd4 67 Bh4! Na7+ 68 Kb7 Nb5 69 Be7 soo eads to zugzwag) 63 Kb4 Nc8 64 Kc5 Ke6 65 Kxc6 h4 66 Bxh4 Na7+ 67 Kb6 Nc8+ 68 Kc7 Na7 69 Bf2 Nb5+ 70 Kc6 ad Back's ufortuate kight must step aside for ite's ast paw to promote. 53 Be2! Kg6 54 Bd3+ Kf6 55 Be2 Botvik's sigature time-gag maoeuvres. 55   Kg6 56 Bf3! N6e7 57 Bg5! 1-0





Zugzwag shatters Back's expectatios. Back's pieces freeze, ike mimes. A dscosoate Broste resiged after a de epy cosequetia 40-mute thik.

: Was his rsigatio prmatur?

 omwhat. H is bustd but cary coud hav payd it out for a og tm, hopg for a accuracy o Botvk's part. But I hav a fg Brost was dprssd at havg b so thoroughy outpayd aftr th adjourmt, ad watd to sav his rgy for th fa gam with th whit pics  which Botv aagd to draw ad hod o to his tit. Game 0 B R Champioship, Moscow 1 emi-lav Defence 1 4 5 2 c4 c6 3 Nc3 N 4 Nf3 e6 5 e3 Nb7 6 B3 Bb4?!

hat we kow ow was a mystery back the. Botvik repeats Euwe's dubious move from Chapter Four (see Game 32). 7 0-0 0-0 8 Bd2 Bd6 9 b3 Qe7 10 Qc2 e5 11 cxd5 cxd5 12 dxe5 Nxe5 13 Nd4 Nxd3 14 Qxd3 Qe5 15 f4 Qe7 16 Racl Rd8 17 Rc2 Bc5 18 Na4 Bxd4! 19 Qxd4 Bf5 20 Bb4 Qd7 21 Rc3 Ne4 22 Rccl b6 23 Rfdl f6! 24 Nc3 Nxc3 25 Rxc3 Be4 26 Qd2 Qg4 27 h3 Qg6 28 Qf2 h5 29 Kh2 as 30 Ba3 b5 31 Bc5 b4 32 Rccl Rdc8 33 Bd4 Bc2 34 Rd2 Be4 35 Rddl Qf5 36 Qe2 Qg6 37 Qf2 a4! 38 Rxc8+ Rxc8 39 bxa4 Qe8 40 Rd2 Qxa4 41 Qh4 Rc2 42 Rxc2 Qxc2 43 Qg3 Qxa2 44 Bxf6

: I raiz that Back has a passd b-paw but is't

Botvik i big troube? His kg

ooks more exposed tha hite's.

 Botv had fors this positio ad was rady with a sick aswr. Ev ftr th combatio, th positio is st probaby draw, but it is Back's sg chac to w i Whit gos wrog. I fact, t's do a xrcis:  ( ): Back pays a itt trick to w matria. How?

 Th qu cbrats hr di vorc with a party.

44  Qxg2+! The tows fok beg to murmur agriy, suspectg the back quee of bewitchig

their beoved kig. She is a amora opportuist who maipuates matters to her favour, as she rushes to fi a power vacuum ad hod sway ad everage over a who oppose her. There are may who fd the charismatic back quee fascatg. However, ite's roya pair are ot amog this group.

45 Qx Bx 46 Bd4 Obviousy forced, sce hite ca resig the kg ad paw edg immediatey after 46 Kxg2?? gxf6. 46  Be4

A ew situatio arises. Back is ow a paw up but faces great tecica difficuties due to the coversio woes voved with opposite-cooured bishops.

: Is't this a dad draw? O xtra paw is

ot usuay eough with bishops of

opposite coours.

 Oy a thradbar fiamt of wlg chacs xist ad Botvik pad to driv aximum bfit from thm. But dad draw? No. Whit wi b forcd to strugg ad fd good movs to achiv his goa.

47 Kg3 Kf7 48 4 Guarateeg the safety of  remag paws. 48  g6 Ditto.

49 Kf2 Ke6 50 Ke2 Kf5 Back fai to make progress if he tries kg etry o the queeside. hite simpy pats his kg o b2 ad waits. 51 Kd2 Kg4 52 Bf6 Kg3 53 Be7 Kh3

54 Bf6?! 54 Bxb4! is a cear path to the draw: 54 ... Kxh4 55 Ke2 Kg4 56 Kf2 h4 57 Kg1 Kf3 58 Bc5 Kg3 59 Bb4 h3 60 Bel + ! Kf3 61 Bd2 Bf5 62 Kh2 Ke2 (the mad kg swirs about  a series of meagess capers, ike a druk at a dace) 63 Bc1  etc.

: Why you th Kotov avoidd this ?

 I thk Kotov saw it d kw it was draw but bivd h had o d to atr cours.

54 000 Kg4 55 Be7 Bf5! The precursor to a crafty idea, his oy shot at w g. Botvk optimisticay hopes to bridge the gap betwee the impossibe ad the possibe. I the temporary respite, Botvk's forces rest ad recharge before the ext taxg batte. 55 ... b3 56 Kc3 Bc2 57 Bf6 Kg3 58 Kb2 Kf3 59 Bd4 g5 60 fxg5 Kg3 61 g6! is draw. 56 Bf6? Now Back is wg. Exceed critica mass ad you risk ahiatio. For the time beg Kotov s mstakey cotet to ay quiescet ad await Botvik's ext thrust. Oce aga Kotov msassesses ad fais to comprehed the magitude of his oppoet's idea. hite sti draws after 56 Bxb4! Kxh4 57 Ke2, as  the ote with 54 Bxb4 above. 56 000 Kf3 57 Be7 b3! 58 Kc3 Be6! Kotov must have expected 58 ... Kxe3? 59 Kxb3 Kxf4 60 Bg5+ Ke4 61 Kc3 d4+ 62 Kd2, which is easiy draw, despite Back's two extra paws.

59 Bc5

·



The positio truy ooks ike a dead draw. After a, White's bishop easiy covers a hs paws, whie his kig hats the b3-passer.

 ( ): This o is dffict. Back ds

two passers, ot oe, to pu off

the wi from this positio. How did Botvk maage ths impossibe-ookg feat?

A Itrfrcdfctio. Back forcs th cratio of a scod passd paw.

59 000 g5!! For the remader of the game, Kotov oses cotro over every phase of his rather dreary desty. Now White is deied his deepest wish, ike Hery VIII, who dreamed of sos but, to his costeratio, was repeatedy haded daughters ad stibor sos by fate, which opey mocked him.

60 fxg5 60 hxg5 h4 61 f5 Bxf5 62 Bd6 Kxe3 63 Kxb3 d4 ws. 60 000 d4+! The briiat pot behid 58 ... Be6!. Botvik correcty diagoses that  order to w, he must reta his b-paw. Back's huk of a bshop heaves up hs scrawy c5couterpart ad rattes hs fae teeth. Through back magic, the bishop cotro both White passed paws o its diagoa, ad at the same time cotro his ow, key b3passer.

61 exd4 Kg3 At og ast, the kg fds a measure of joy previousy deied him. Back is actuay dow a paw ow, yet wg! From this pot White ca oy erect dsprited resstace.

62 Ba3 Freeg his ow kg, but it is too ate. Both moarchs joste for positio, oe seekg roads, the other fightg for surviva. 62 000 Kxh4 63 Kd3

Or 63 g6 Kg3 64 Bd6+ Kg4 65 g7 h4 ad ite has o defece to . . . h4-h3, . . . Kf3g2 ad ... h3-h2. 63 000 Kxg5 64 Ke4 h4 65 Kf3 Bd5+ 0-1

.

.

Nothg is more pafu tha wakg away empty-haded from a og abour.

: How dos Back w i Whit pacs his kg o h2 ad simpy waits?

 It's oy a attr of tm bfor Whit must submit to coqust by attritio: 66 Kf2 Kf4 67 Kg K4 68 2 B6! (oh, o you do't! ) 69 Bb2 h3 70 Ba Kd3 71 d5 Bd 7 72 d6 Kc2 73 Kg (th dazd kg wadrs aissy, ik a asia victm whos tir past is a xpgd void  h is rgatd to his w, humb post, shuffg aissy d awaitg his fat) 73 . . . b2 74 Bxb2 Kxb2 ad Back ws. fortaty for Kotov, his oppot ras with th corrct coord bishop to promot th h-paw.

It's sometimes difficut to reta oe's faith  the deity vsits heavy tribuatio upo the faithfu. I this game Caissa smied oy upo Botvk, whie igorg Kotov's fervet peas for the draw.

Index of Games 1936 Moscow 1956 Mote Carlo 1968 USSR Absolute BotvkM-Alatosev 1934 USSR AVRO Wo Wo AVRO Moscow 1935

1941 Moscow 1931 The Netherlas 1938 Moscow 1951 Moscow 1951 The Netherlas 1938

Moscow 1936 USSR Moscow 1952 Worl The Worl The Palma e Malloa 1967 USSR Moscow 1940 Worl Mote Carlo 1968

1948 1948

Moscow 1963

USA v s U SS R rao match 1945 USS R Absolute Worl USSR

Mea

The Moscow 1944

1941 1948

Wo vs USSR 1970 Palma e Malloa 1967 Alekhe Moscow 1956 Moscow 1952 Worl Moscow 1963 USSR Moscow 1927 USA vs USSR Moscow 1946 Moscow 1952

Tabe of Contents About the Author Itrod uctio 1 Botvno the Attack 2 Botvno Defece 3 the Eemet 4 Botvno Imbaaces 5 Botvno 6 Botvno Idex of 1936 Moscow 1956 Mote Caro 1968 1941 Botvi M-Aatortsev.

1934

USSR AVRO Word Word Botvi AVRO Botvi M-Chekhover. Moscow 1935

Moscow 1931 The Netherads 1938 Moscow 1951 Moscow 1951 The Netherads 1938

Moscow 1936 USSR Moscow 1952 Word The Pama de Maorca 1967 Botvi M-Levefish. USSR Moscow 1940 Word Mote Caro 1968

Botvi M-Vidmar.

Keres.P-Botvi

1948 Moscow 1963

1936

USA vs. USSR radio match 1945 Moscow 1947 USSR Absoute

1941

"

dce About the Author Bibiography Introduction 1 Botvinnik on the Attack 2 Botvinnik on Defence 3 Riding the Dynaic Element 4 Botvinnik on Exploiting Imbalances 5 Botvinnik on Accumulating Advantages 6 Botvinnik on Endings Index of Opponents Alekhine.A-Botvinnik.M, Nottingham 1936 Averbakh.Y-Botvinnik.M, Training match, Moscow 1956 Benko.P-Botvinnik.M, Monte Carlo 1968 Bondarevsky.I-Botvinnik.M, USSR Absolute Championship, Leningrad/Moscow 1941 Botvinnik.M-Alatortsev. V, Leningrad 1934 Botvinnik.M-Alatortsev.V, USSR Championship, Moscow 1931 Botvinnik.M-Alekhine.A, AVRO Tournament, The Netherlands 1938 Botvinnik.M-Bronstein.D, World Championship (23rd matchgame), Moscow 1951 Botvinnik.M-Bronstein.D, World Championship (7th matchgame), Moscow 1951 Botvinnik.M-Capablanca.J, AVRO Tournament, The Netherlands 1938 Botvinnik.M-Chekhover.V, Moscow 1935 Botvinnik.M-Euwe.M, World Championship Tournament, The Hague /Moscow 1948 Botvinnik.M-Fischer.R, Varna Olympiad 1962 Botvinnik.M-Flohr.S, Moscow 1936 Botvinnik.M-Furman.S, Training match, Moscow 1961 Botvinnik.M-Keres.P, USSR Championship, Moscow 1952 Botvinnik.M-Keres.P, World Championship Tournament, The Hague/Moscow 1948 Botvinnik.M-Larsen.B, Palma de Mallorca 1967 Botvinnik.M-Levenfish.G, USSR Championship, Moscow 1940 Botvinnik.M-Petrosian.T, World Championship (14th matchgame), Moscow 1963

4 6 7 15 81 127 186 230 291 355 131 339 120 150 28 20 304 342 324 45 32 156 112 230 255 70 318 281 140 291