Professional Documents
Culture Documents
For further details for Batsford chess titles, please write to Batsford Chess
Books, 583 Fulham Road, London SW6 5BY.
Batsford Chess Opening Guides
John Emms
Bibliography 8
Introduction 9
Books
Encyclopaedia of Chess Openings vol.C, Sahovski Informator 1997
Batsford Chess Openings 2, Kasparov & Keene (Batsford, 1989)
The Complete French, Psakhis (Batsford, 1992)
Winning with the French, Uhlmann (Batsford, 199 1)
Mastering the French, McDonald & Harley (Batsford, 1997)
Play the French, Watson (Cadogan, 1984 and 1986)
The Modern French Tarrasch, Gufeld (Cadogan, 1996)
Periodicals
Informator
ChessBase Magazine
New In Chess Yearbook
British Chess Magazine
Chess Monthly
INTRODUCTION I
9
The Fre n c h Ta rra s c h
10
In troduction
can choose between attacking Black's Black. His queen is doing a marvellous
weak backward e6-pawn, or piling up job on g7, where it protects the black
on the c-file. In contrast, Black has no kingside, attacks the weak d4-pawn
constructive plan and can only sit and and prepares a pawn assault starting
wait, while White leisurely builds up with 35 ... g4! See Game 14 for the rest
the pressure. Notice that all of White's of this very instructive encounter.
pieces stand on dark squares.
Wells-Fries Nielsen
Copenhagen Open 1995
Marjanovic-Timman
Sarajevo 1984
see following diagram
11
The Fre nch Tarra s c h
The ... e6-e5 break has constant problems with his d- and
In many Tarrasch positions the move b-pawns.
... e6-e5 enables Black to free his pieces
and thereby solve most of his prob
lems immediately.
Bialas-Uhlmann
Leipzig 1951
Emms-Poulton
Hastings 1997
see fol/o.ving diagram
Here's an example of Black failing to In this position Black has the ex
solve his problems, even after achiev traordinary move 10 ... g5!, which has
ing the desired advance. After 14 ... e5 indeed become the main line here. See
15 dxe5 tbxe5 16 tbxe5 "ifxe5 17 "ifb3 Games 32-33 for more details on this
White retains a more comfortable po variation.
sition. In this case Black has no attack Here's an even more extreme ex
ing chances against the white king, but ample.
12
I ntro duc ti on
Adams-Luther
Oakham 1990
3 ... c5
13
The F re n c h Tarra s c h
importance of this is often exagger- while having a few guidelines for these
ated. Black also holds some trump sorts of positions.
cards.
14
I n t ro duc tion
In this position Black's strategy has the above diagram there are three pos
failed. Two sets of minor pieces have sibilities, but only one gives White
been eliminated and White has the e4- chances of an advantage. White can
and c4-squares firmly under his con exchange a minor piece with 1 i.xc6
trol. More importantly, however, the bxc6. However, that change in the
d-pawn is chronically weak and is pawn structure favours Black, as the
ready to be picked off at any moment. dS-pawn is no longer isolated. An
other idea is 1 i.xe7, but after
1...i.xe7 Black gets control of the dark
squares and his bishop can be used to
support a later . .. d5-d4. The best solu
tion is 1 i.h4!, intending i.g3, offering
a trade of those dark-squared bishops,
which would eliminate Black's best
piece and increase White's control
over the dark squares. Throughout the
book you'll find me harping on con
tinuously about the importance of
these colour complexes. I apologise in
Part of White's strategy against the advance, but their importance can
IQP is to exchange the right pieces. In never be underestimated!
15
I CHAPTER ONE I
3 '2Jf6:
0-0
. . .
16
3 . . . l'Df6: Main Line with 1 1 . . . 0-0
17
The Fre n c h Ta rra s c h
b) 19 f4 tbg4 (19 ... tbxd3!? also de <it>xa2 5 1 !Ixb4 <it>al 52 <it>d2 h2 53 !Ih4
serves attention) 20 'ifd2 d4 2 1 i.e4 <it>b2 54 !Ixh2 a2 and finally the draw
tbe3 22 l:.f3 i.g4 23 �f2 'iff6? (I like becomes inevitable.
23 ... i.fS! here, as after 24 i.xfs 'ifdS+
25 tbf3 tbxfS the exchange of bishops
emphasises the weaknesses around the
white king) 24 l:.cl �f8 25 tbe2 'ife6
26 tbg3 'ifh6 27 tbfl ttJfs 28 <it>g1 'ifh4
29 'ifd3 'ifd8? 30 h3! 1-0 Hunt
Hagarova, Yerevan Women's Olym
piad 1996.
1 9 . . .d4 20 .ie4 .ie6 21 �c 1 •h4 22
f3 �dB 23 �e 1 •ta
Game 2
Kramnik-Ulibin
USSR Championship 1991
1 e4 e6 2 d4 d5 3 lt:Jd2 lt:Jf6 4 e5
lt:Jfd7 5 c3 c5 6 .id3 lt:Jc6 7 lt:Je2
cxd4 8 cxd4 f6 9 exf6 lt:Jxf6 1 0 lt:Jf3
.id6 1 1 0-0 0-0 1 2 .if4 .ixf4 1 3
lt:Jxf4 lt:Je4 1 4 lt:Je2 �xf3 !? 1 5 gxf3
Black has completed his develop lt:Jg5 1 6 'iii>h 1 e5! 1 7 dxe5 lt:Jxf3 1 8
ment and his pieces stand harmoni .ixh7+ 'it>h8!
ously. All in all, he has a very reason 18 ... @xh7 19 'ifd3+ <it>g8 20 'ifxf3
able position. tbxeS 21 'it'g3 is good for White.
24 b3 .id5 25 b4! d3 26 .ixd5+ 1 9 lt:Jg 1
�xd5 27 •b3 •t1 28 �ed 1 g5! 29
�c5 �xc5 30 bxc5 lt:Jd4 31 "ii'xf7+
�xf7 32 lt:Jh3! lt:Jdxf3 33 lt:Jf2 d2 34
�g2 �e6 35 lt:Je4 �d5 36 lt:Jxd2
lt:Jxd2 37 �xd2+ �xc5 Yz Yz -
18
3 . . . l'Df6: Main L ine with 1 1 . . . 0-0
19
The Fren c h Ta rra s c h
20
3 . . . 4J f6 : Main L ine with 1 1 . . . 0-0
'lt>f4 'ifg2 25 ttJg3 ttJb4! 26 'ifb3+ necting White's queen from the de
ttJdS+ 27 'lt>g4 'lt>h8 28 .l::!.ae l l::.f 8!, and fence of the vital d-pawn.
White's king has slipped into a mating 27 l:tc 1
net. Novotny-Nilsson, European After 27 a3 White gets hit with
Team Correspondence Championship 27... ttJb4! (anyway) as 28 .5lxd7 'ifxd4+
199 1, concluded 29 h4 l:I£4+ 30 'lt>hs 29 'lt>e2 'ifd3+ 30 'lt>f2 .l::!.c2+ 3 1 'ifxc2
.l::!.fS+ 3 1 'lt>g4 g6 32 .l::!.e8+ 'lt>g7 33 ltJxc2 is good for Black.
'ifxb7+ .l::!.f7 34 'ifxf7+ 'lt>xf7 35 .l::!.e S 27 . . . lleS! 28 llfe 1 tLlb4! 29 J:te5
h5+ 36 'lt>gS 'iff3 37 'lt>h6 'iff4+ 0-1, as 29 .l::!.xe8+ .5lxe8 30 .5lxe8 'ifxd4+ 3 1
after 38 .l::!.g S 'ifxd4 39 .l::!.xg6 ttJf4 the 'lt>e2 'ifxb2+ 3 2 'lt>dl 'ifd4+ is also a
rook goes or it is mate. draw by perpetual check.
20 . . . exf4+ 2 1 'iti>d2 29 . . . llxe5 30 dxe5 'il'b6 3 1 i..xd7
2 1 'lt>xf4 'ifh6+ 22 'lt>f3 .5lh3 23 .tf5 'il'f2+ 32 'iti>d 1 'il'f1 + 1h -1h
.l::!.f8 occurred in De Souza-Ferraro,
Zurich 1990, a game that White went Game 4
onto win. Instead of 23 ... :f8, Black Komarov-Barsov
should play 23 ...'iff6!, hitting the d4- Reims 1994
pawn. Then 24 %:!.e l ltJxd4+ 25 'ifxd4
'ifxd4 26 .5lxh3 .l::!.f8+ and 24 'ifd3 ltJe7! 1 e4 e6 2 d4 d 5 3 tLld2 tLlf6 4 i..d 3
21
The Fre n c h Ta rra s c h
22
3 . . . li:if6: Main L i ne with 1 1 . . . 0-0
23
The Frenc h Ta rra s c h
24
3 . . . li:i f6: M a in Line with 1 1 . . . 0-0
1 e4 e6 2 d4 d5 3 l2Jd2 l2Jf6 4 e5
l2Jfd7 5 c3 c5 6 .td3 l2Jc6 7 l2Je2
cxd4 8 cxd4 f6 9 exf6 l2Jxf6 1 0 0-0
i..d6 1 1 l2Jf3 0-0 1 2 .tf4 i..xf4 1 3
l2Jxf4 l2Je4 1 4 'ii'c 1 l2Jg5 1 5 l2Jxg5
'ii'x g5 1 6 i..x h7+!?
25
The Fre n c h Ta rra s c h
26
3 . . . fiJf6: Main L in e with 1 1 . . . 0-0
.l::!.e2+ 34 @fl :xb2 and it is Black who .l::!. fcl . This is far from the end, how
is doing the pushing. ever, as Black can play 18 ...:xf3 19
30 . ..l:th3+ 3 1 �2 .txd5 1h -1h gxf3 ltJxd4 with a very messy position.
32 :dxdS ttJxdS 33 !hds .l::!.xh2+ 34 18 llad 1 ..id7 19 'il'a3 l2Jxd4 20
'lt>g3 l::.xb2 leads to a totally drawn l2Jxd4 'il'xd4 2 1 ..ixg6 'il'f6 22 ..ib 1
rook and pawn endgame. ..ib5 23 ..id3
Game l
Henao-Sequera
M erida 1992
1 e4 e6 2 d4 d5 3 l2Jd2 l2Jf6 4 e5
l2Jfd7 5 c3 c5 6 .td3 l2Jc6 7 l2Je 2
cxd4 8 cxd4 f6 9 exf6 l2Jxf6 1 0 l2Jf3
.td6 1 1 0-0 0-0 1 2 .tf4 ..ixf4 1 3
l2Jxf4 l2Je4 1 4 l2Jh5
The knight heads for g3, where it
will confront its black counterpart on
e4. This is a perfectly logical, albeit 23 . . . ..ic6!
rather time-consuming plan. Ulibin's suggested improvement
1 4 . . . g6 1 5 l2Jg3 l2Jxg3 1 6 hxg3 over 23 ... .5lxd3, which he labels as a
' strategic howler'. Spasov-Ulibin, Tun
ja 1989, continued 24 .l::!.xd3 :ac8 2S
.l::!.f3 'ifg7 26 %:!.e l :c6 27 'ifb4 bs 28 b3
.l::!. fc8 29 :fe3 l:Ie8 30 .l::!.e S and Black's
shaky king position had become the
overriding feature of the position. The
idea behind 23 ... .5lc6 is to preserve the
bishop in order to support the central
pawn mass.
24 'il'b3 '1ti>h8 25 ..ib1 e5 26 .:td2 llf7
27 lle 1 llg8 28 'il'c3 .J:!.g5 29 .tc2 h5
In contrast with the Spasov-Ulibin
Now White threatens .tbs, ex game, the principal features here are
changing on c6 and establishing a mas Black's centre and active pieces. Black
sive bind on the dark squares. Need now has no problems at all.
less to say, Black has to remain very 30 ..ib3 d4 31 'il'c4 l:te7 32 'il'c5
active. .:teg7 33 .td1 h4 34 g4 h3 35 gxh3
1 6 . . .'il'b6! 1 7 'il'a4 a6
see following diagram
17 ...'ifxb2!? is meant to be unplay
able due to 18 .tbs, threatening to 35 . . .llxg4+ 36 hxg4?
trap the black queen with %:!.ab 1 and After 36 .5lxg4 'iff3 37 @fl Black
27
The Fre n c h Tarra s c h
1 4 . . . g5
The most ambitious reply to 14 g3,
but perhaps not the best. Other ideas
include:
a) 14 ...'ii'f6 (perhaps the most testing
response) 15 h4 h6 16 .5lxe4 dxe4 17
36 . . .'il'h4 37 'il'xc6 bxc6 38 llxe5 ttJe5 l:.d8 18 ttJxc6 bxc6 19 'ifc2 g 5 20
llh7 39 .J:!.e8+ ..ti>g7 40 ..if3 'il'h2+ 41 hxg5 hxg5 2 1 ltJe2 'iff3 22 'ii'd2? 'lt>f7
�1 'il'f4 23 'ifxg5 :h8 24 'iff4+ 'lt>e7 25 'ifxf3
A vigorous technical phase lies exf3 26 ltJf4 .l::!.h6 27 ttJd3 and White
ahead, but White's resistance was not resigned in Kotronias-Ulibin,
up to much. Chalkidiki 1992. Ulibin gives 22 ltJc3
42 lld3 'il'c 1 + 43 l:te 1 'il'c4 44 lled1 as an improvement for White and as
c5 45 b3 'il'b5 46 a4 'il'b6 47 .:te 1 sesses 22 ... .5la6 23 :fe l .td3 24 'ifd2
'ill'd6 48 lle8 'ii'g6 49 .:te4 .:th8 50 'iff5 as unclear. There is plenty of
b4? l:!.e8! 0-1 scope for home analysis on this posi
After 51 �xe8 'ifxd3+ 52 'lt>g2 cxb4 tion!
the two passed pawns prove to be de b) 14 ...ltJg5 15 ltJe5 ltJxe5 16 dxe5
c1S1ve. l::.xf4!? 17 gxf4 ltJh3+ 18 'lt>h 1 ltJxf4 19
.l::!. g l .5ld7 20 'it'g4 'iff8 21 .l::!.g3 .5lc6 22
.l::!.dl d4+ 23 'lt>gl 'iff7 and here the
Game 8 players agreed a draw in Geller-Zi.iger,
Stigkin-Basin Bern 1988, just when the position was
St Petersbu rg 1992 getting interesting. Clearly Black has
substantial compensation for the ex
1 e4 e6 2 d4 d5 3 tLld2 tLlf6 4 e5 change in the final position, so White
tLlfd7 5 ..id3 c 5 6 c3 tlJc6 7 tLle2 should look for improvements earlier
cxd4 8 cxd4 f6 9 exf6 tLlxf6 1 0 0-0 on .
..id6 1 1 tLlf3 0-0 1 2 ..if4 .txf4 1 3 1 5 tLlh5 e5 1 6 tlJxe5! tlJxd4
tLlxf4 tlJe4 1 4 g3 16 ... .5lh3 falls short after 17 .5lxe4
28
3 . . . lll f 6: Main L in e with 1 1 . . . 0-0
29
Th e Fre n c h Tarra s c h
30
3 . . . ll'if6: Main L i n e with 1 1 . . . 0-0
After 24 ... tDe3 25 .l::!.f 3! tDxfs 26 .l::!.afl possibilities. Other ideas include:
lbd6 2l :f8+ 'lt>gl 28 h4 White has a a) 12 h3 (this quiet line should cause
winning position. The text move Black no problems whatsoever)
22 .....tfs leads to a very unusual posi 12 ... .tdl 13 .5le3 'ii'el 14 1:%.cl a6 1 5 a3
tion. White has four connected passed 'lt>h8 16 %:!.el .5le8 ll t2lg3 tDhs 18
pawns, but they are all well blockaded tbxhs .5lxh5 19 g4 .5le8 20 'lt>g2 'ii'f6 2 1
and in fact it is Black's lone passed d .5lg5 'iifl 2 2 .th 1 lbxd4! and Black had
pawn that wins the day. won a pawn in Zaichik-Panchenko,
23 f4 �7 24 llfd 1 ..ti>e6 25 l:td2 Lvov 198l, as 23 lbxd4 'ii'xf2+ 24 'lt>h l
llh8! 26 llc1 llh4 27 l:tf2 d4 28 ..ti>f1 'ifh2 is mate.
ttJb4 29 llc 7 tLld3 30 llxb 7 b) 12 lbc3 a6 (12 ... 'ifb6 transposes to
Giving up an exchange, but after 30 Chapter 2, Games 16-18) 13 .5le3 (13
l:If3 tDxf4 White's position collapses. .5lg5!?) 13 ... ..tdl 14 %:!.cl 'ii'el (naturally
30 . . .tLlxf2 3 1 ..ti>xf2 .:txf4+ 32 ..ti>e 1 cl and b6 are also possible squares for
'1ti>xe5 33 llxa7 '1ti>e4 34 lle7+ ..ti>d3 the queen, but here Black opts for the
35 lla7 ..ti>e3 36 b4 d3 0-1 kingside) 15 l:e l lbg4 16 .5lg5 'ii'fl ll
The d-pawn will soon advance to h3 h6!? 18 .5lh4 tDf6 19 .5lg3 .5lxg3 20
promotion. fxg3 'iie l 2 1 a3 'ifd6 22 'lt>h2 .5le8 23
'ife2 .tbs 24 'ifxe6+ 'ii'xe6 25 l:Ixe6
Game 10 .5lxf3 26 gxf3 tDxd4 2l %:!.el l;tfl with
Rozentalis-Hergott an equal endgame in Rovid-Ulibin,
Montreal 1995 Cappelle la Grande 1994.
c) After the tricky 12 %:!.e l!? Black's
1 e4 e6 2 d4 d5 3 tLld2 tLlf6 4 e5 best bet looks to be a transposition to
lLlfd7 5 i..d 3 c5 6 c3 tlJc6 7 tLle2 Chapter 2 with either 12 ...'ifb6 or
cxd4 8 cxd4 f6 9 exf6 tLlxf6 10 tLlf3 12 ...'ifcl, as both 12 . . ...tdl and
..td6 1 1 0-0 0-0 1 2 i.. g 5 12 ...'ife8 may be answered by 13 .tf4,
which gives White an edge.
1 2 . . . 'il'e8
The only other independent idea is
12 ... ..tdl!?, keeping Black's options
open. After 13 .5lh4 .5le8 14 'ifb 1 h6
15 .5lg6 .5lxg6 16 'ii'xg6 'ii'e8 ll 'ifxe8
:axe8 18 .5lg3 lbe4 19 .5lxd6 tDxd6 20
.l::!.fdl lbe4 21 t2lg3 lbxg3 22 hxg3 .l::!.c8
23 .l::!. d2 @fl 24 %:!.el 'litel the players
agreed a draw in T.Horvath
Kindermann, Germany 1996. Once
again Black also has 12 ... 'i'b6 and
This is probably the most signifi 12 ...'ifcl.
cant alternative to 12 .tf4, although it 1 3 i..f4!
does contain many transpositional A clever switchback! Rozentalis be-
31
The Fre n c h Ta rra s c h
1 5 . . . ..id7
Aggressive measures only backfire
for Black. For example, 15 ... gS 16
.ixe4 .l:r.xf4 17 .ixh7+ 'lt>xh7 18 lDxgS+
wins, while 16 ...gxf4 17 .id3 leaves
Black's pawn structure in a real mess.
Black's 'extra move' ... 'ife8 also means
that the positionally desirable ...lDgS is
impossible. All in all, it would appear
that White can retain a significant After 28 ....idS! Black's active pieces
edge. ensure good drawing chances.
1 6 'il'e3 l2Jb4 1 7 ..ixe4?! 29 l2Je4! l:te2 30 l2Jg5+ \tie 7 3 1 l:Z.xe2
After this move Black obtains some l:!.xe2 32 ..ti>f 1 ..ig4
unexpected counterplay. Rozentalis After 32 ...l::.c2 33 %:!.e l or 32 ... .l::!.eS 33
gives 17 tDeS tDxd3 18 tDfxd3 .ibS 19 tDxe6 'lt>xe6 34 l:Ie 1 White can achieve
.l:r. fel .ixd3 20 tDxd3 as an improve a won king and pawn endgame. Now,
ment. however, Black loses the exchange.
1 7 . ..dxe4 1 8 'il'xe·4 ..ic6! 33 h3 lle5 34 f4 1 -0
32
3 . . . li:Jf6: Main Line with 1 1 . . . 0-0
Summary
Despite rigorous tests, the 12 .if4 .ixf4 13 ltJxf4 ltJe4 14 ltJe2 l:txf3 exchange sac
rifice continues to annoy White, although it is clear that both players need to
know their stuff before they venture into these critical lines. White players
wanting an easier life will probably prefer 14 °ifcl, which is certainly a super
solid option. It will be interesting to see whether 14 ltJhS or 14 g3 become more
popular.
1 2 i..f4
12 .igS - Game10
1 2 ... hf4 1 3 llJxf4 llJe4 (D)
13 ..."ifd6 - Game9
1 4 l2Je2
14 °if cl ltJgS 15 ltJxgS °ifxgS (D)
16 ltJe2 Game5 -
16 .ixh7+ Game6 -
14 ttJhs Game7
-
14 g3 Game8
-
16 ...ttJh3+ Game4 -
1 8 .. llJfxeS
. Game1
-
33
CHAPTER TWO I
3 ...'2Jf6: Main Line
with 11...'i/c7 and 11 ...'ii'b 6
34
3 . . . ll'i f6: Main Line with 1 1 . . . 1kc7 a n d 1 1 . . . 1k b 6
1 2 . . 0 -0 1 3 i.. h 4
. .:tae8 20 h3 .:te 7
Again this is the most fashionable 20... .ibS 21 'iib4 .ixd3 22 'iixb6
move. 13 %:!.cl and 13 ltJc3 a6 are the axb6 23 .l::!.xd3 still leaves White with
subjects of Games 13 and 14 respec an edge, due to the pawn weaknesses
tively. in the black camp.
1 3 . . .e 5 2 1 llfe 1 .:tfe8 22 .:txe7 .!Ixe7 23 i..c2
We shall look at 13 ...ttJhS in Game .tb5 24 'il'g5 h6 25 'il'd2 a5 26 .tb1
12. .:td7
14 dxe5 l2Jxe5 1 5 l2Jxe5 i.. x e5 1 6 If 26 ... d4 then not 27 'iixd4 'iixd4
.tg3 i.. x g3 1 7 l2Jxg3 28 lhd4 %:!.el+, but 27 .ifs! and Black
has problems defending the d4-pawn.
27 l2Jf5 l2Je4 28 'il'd4 'il'e6
28 .. .'iixd4 29 ltJxd4 .ia6 30 ttJb3
wins a pawn for White, as e4 and a5
are attacked.
29 l2Je3 i..c6 30 a3 'il'f6 3 1 f3 'il'xd4
32 llxd4 l2Jf6 33 �2
35
The Fre n c h Ta rra s c h
Game 12
Yemelin-Kostenko
S zeged 1994
1 e4 e6 2 d4 d5 3 l2Jd2 l2Jf6 4 e5
l2Jfd7 5 c3 c5 6 .td3 l2Jc6 7 l2Je2
cxd4 8 cxd4 f6 9 exf6 l2Jxf6 1 0 l2Jf3
..id6 1 1 0-0 'il'c7 1 2 ..ig5 0-0 1 3
.th4 l2Jh5
45 b5! ..ixb5 46 l2Jxd5 ! The major alternative to 13 ... e5, al
46 .l::!.xd5 l:txd5 47 .5lxd5+ @e7 48 though the uncommon 13 ... .5ld7 also
.5lxb7 also maintains reasonable win looks quite playable.
ning chances, but I was tempted by a 1 4 'il'c2
mating net. The most questioning reply. Other
46 . . . 'i;f5 47 llb2!? ..ic6 48 l2Jf6 llg7 moves do nothing to prevent Black's
49 lle2 l2Jc5? standard plan of ... g7-g6 and .. . °ifg7.
49 ...@f4 was the last chance. Now
the web closes around the black king.
50 'i;g3 ! ..ib5 5 1 ..ib 1 + ..id3
1 4 . . . h6
The natural 14 ...g6 falls short after
15 .5lxg6! and now:
52 f4! a) 15 ... hxg6 16 "ifxg6+ ltJg7 17 ltJg5
Needless to say, I was very satisfied and wins.
with the conclusion of this game, but b) 15 .. Jlxf3 16 .txh5 .5lxh2+ 17
in fact it turns out White has an alter @h l .l::!.f5 18 .5lg4 .l::!.f8 19 g3 and the
native finish with 52 .5lxd3+ ltJxd3 53 bishop is lost.
36
3 . . . ll'i f6: Main Line with 1 1 . . . 1kc7 a n d 1 1 . . . 1kb6
1 5 i.. g 6 l2Jf4
After this move White tends to ob
tain a clear advantage, so perhaps
Black players should look into
15 ...llxf3!?, which hasn't been wit
nessed much but is a trickier move.
After 16 gxf3 .5lxh2+ 17 'ifi>hl ltJf4 18
ltJg3 Black has two ways to play:
a) Akopian gives 18 ...ttJxg6 19 'ifxg6
.5lxg3 20 .5lxg3 'iff7 2 1 'ifxf7+ 'iti>xf7 22
nfdl with only a minute advantage
for White in the endgame. Indeed
Black's position does seem quite diffi
cult to breach.
b) Altgelt-Stamenkovic, Sofia 1994, Game 13
saw instead 18 ... .5lxg3 19 .5lxg3 .5ld7 Smagin-Dolmatov
20 'ifd2 l:.f8 with some compensation USSR Championship 1986
for the exchange. Indeed, Black went
on to win this game. 1 e4 e6 2 d4 d5 3 l2Jd2 l2Jf6 4 e5
1 6 l2Jxf4 i..xf4 1 7 .:tad 1 l2Jfd7 5 i..d3 c5 6 c3 l2Jc6 7 l2Je2
Black's predicament lies with the cxd4 8 cxd4 f6 9 exf6 l2Jxf6 1 0 0-0
shaky light squares around his king. i..d6 1 1 l2Jf3 'il'c7 1 2 i..g5 0-0 1 3
White has a very simple plan of llc1
switching the roles of bishop and An alternative way to proceed. The
queen along the b 1-h7 diagonal and objective is to dissuade ... e6-e5 before
playing for checkmate. It is difficult to proceeding with the .5lh4-g3 plan. Of
find a suitable antidote to this idea. course the tempo spent achieving this
1 7 . . . i..d7 1 8 'il'e2! a6 1 9 .:tfe 1 l2Jd8 idea gives Black more time to start a
20 l2Je5 i.. x e5 21 dxe5 llc8 22 b3 counterattack.
i..e8 23 i.. b 1 g5 1 3 . . .l2Jg4!
Extremely committal, but Black Cutting across White's main goal.
feels obliged to cover h7. Now 14 h3 allows 14 ...nxf3!, when 15
24 i..g 3 'il'g7 25 'il'd3 l:!.f7 26 'il'd2 gxf3 ltJh2 and 15 hxg4 .l::!.f7 are both
.:tfc7 fine for Black.
1 4 l2Jg3 g6 1 5 l2Jh4
see following diagram
15 tiJd2!? is a fairly new move of
2 7 f4! Tiviakov's which certainly deserves
The start of a decisive breakthrough attention. After 15 ...ttJf6 16 ttJb3 'it'g7
27 . . .i.. g 6 28 f5! exf5 29 'il'xd5+ '1ti>h7 17 'ifd2 .5lb4 18 'ife3 ltJg4 19 'ife2 ltJh6
30 e6 f4 31 e7 llxe7 32 i..xg6+ 20 .tbs .5ld7 2 1 .5lxc6 bxc6 22 ttJcS
..ti>xg6 33 'il'd6+ 'il'f6 34 llxe7 fxg3 .5lxc5 23 .5lxh6 'ifxh6 24 :XcS Black's
35 'il'xf6+ ..ti>xf6 36 lle8 1 -0 bad bishop condemned him to a pas-
37
The F re n c h Tarra s c h
38
3 . . . 4J f6 : Main L ine with 1 1 . . . • c l and 1 1 . . . • b 6
1 5 .:tc 1
1 5 'ifc2 can b e answered with
15 ... g6! here, as 16 .5lxg6? runs into the
surprising 16 ... ltJf4! 17 .5ld3 ltJb4 and
Black wins a piece. This is one of the
disadvantages of 12 ltJc3.
1 5 . . . g6 1 6 i.. b 1
16 ttJa4 .td7 17 ttJcs lhf3 ! 18 'ifxf3
ltJxd4 19 'it'g4 .5lxc5 occurred in Griin
feld-Shachar, Israeli Championship
1994. Here White's best is 20 'ifxd4, While White is groping for an active
but after 20 ... 'ifxh2+! 21 'lt>xh2 .5lxd4 plan, his opponent can steadily im
Black is certainly no worse. prove his position before expanding
1 6 . . .'il'g7 1 7 'it>h 1 ? ! on the kingside. Already Black's posi
This seems a little slow. White can tion is easier to play.
also try: 2 1 llc3 h6 22 i..g3
a) 17 ttJa4 .l::!.xf3!? (17...l:tb8!?) 18 gxf3 An admission of defeat. White ex
.td7 19 ttJcs .l::!. f8 20 .5lg3 .5lxc5 21 changes his dark-squared bishop, but
dxcS 'ifxb2 2 2 l:te l 'iff6 2 3 'lt>hl ltJd4 only gets a knight in return - a poor
24 .5le5 'ifxf3+ 25 'ifxf3 ttJxf3 was swap.
clearly better for Black in Womacka 22 . . . lbxg3+ 23 lbxg3 llf4 24 lbe2
Pahtz, East German Championship ll4f6 25 lbg3 i..e8 26 llce3 i..f7 27
1985, but I think that White's play can ..ti>g 1 i..g8 28 'il'd3 i..f4 29 ll3e2 'il'f7
be improved upon. 30 b4 g5 31 h3 i..c7 32 lbf 1 i..b6 33
b) 17 .5lg5!?, returning to the c1-h6 lb 1 d2 'il'g7 34 lbb3 .:tf4 35 lld 1 g4!
diagonal, is an interesting idea. The This thrust represents the culmina
point is that with the knight on hS, tion of Black's strategy. The d4-pawn
the bishop has no real future on h4, so is very sensitive, while Black also
it returns to its original diagonal. I dominates proceedings on the king
believe that Karpov was the first to side.
39
The Fre n c h Tarra s c h
, 3 . . ..td7
13 ....txf4 14 ttJxf4 'ifb6!? is a differ
ent idea. Great complications may a) 17 lle 1 a6 18 ltJeS ltJxeS 19 .l::!.xeS
40
3 . . . lll f6 : Main Line with 1 1 . . . 1ic7 a n d 1 1 . . . 1i b 6
'i'Vb6 2 0 :c 2 lbg7! (the knight per �xf7 28 l:l.xe6 'ii'xe6 29 hxg3 �98
forms a very useful defensive role 30 'ii'd 3 'ii'xg4
here) 2 1 'ii'g4 .tbs 22 lbxbS axbS 23
b3 :ac8 24 :ce2 :c3 2S :se3 b4 26
h4 'ii'd8 27 'ii'gS 'ii'f6 28 'ii'xf6 :xf6 29
.tbs :c7 30 'it>g2 'it>f7 31 :d2 'it>e7
with an equal position in Dvoirys
Dolmatov, USSR Championship
1986.
b) Psakhis recommends 17 .tbs,
giving White a small, secure advan
tage, but the sacrifice 17 ... :Xf3! is very
unclear. Indeed, after 18 'ii'xf3 lbxd4
19 'ii'd3 lbxbS 20 lbxbS 'ii'a6 2 1 a4
'ii'xa4 I prefer Black, who has two The knight can be picked up at
pawns and some light-squared control Black's leisure, while the most White
for the exchange. In Malevinsky can hope for in return is a pawn.
Komarov, Leningrad 1989, White Dolmatov, renowned for his immacu
tried 1 8 i.xc6, only to be shocked by late technique, doesn't let it slip.
18 ... :xc3! 19 i.xd7 :xc l 20 i.xe6+ 31 �f2 'ii'e 6 32 'ii'c2 g6 33 'ii'c5
'ii'xe6 2 1 'ii'xc 1 :cs and Black already 'ii'f7+ 34 �g2 �xh7 35 'ii'xc6 h5 36
had a winning position. b4 d4 37 a4 'ii'e7 38 b5 d3 39 'ii'd5
1 5 . . . h6 1 6 iLxd6 'ii'xd6 1 7 ll'ih7 l:l.f3! 'ii'e 2+ 40 �h3 'ii'f 1 + 41 �h2 'ii'f2+
1 8 iLb1 42 �h 1 d2 0-1
Or 1 8 'it>g2 :fs! 19 lbc3 g6 and the
knight on h7 is dead and buried. Game 16
1 8 . . . e5 1 9 l:l.xc6 bxc6 20 dxe5 'ii'x e5 Nijboer-Farago
21 ll'id4 lU7 22 l:l.e 1 Dieren Open 1988
1 e4 e6 2 d4 d5 3 ll'id2 tl'if6 4 e5
ll'ifd7 5 c3 c5 6 il.d3 ll'ic6 7 ll'ie2
cxd4 8 cxd4 'ii'b6 9 ll'if3 f6 1 0 exf6
ll'ixf6 1 1 0-0 il.d6
41
The Fre n c h Ta rra s c h
42
3 . . . tll f6: Main Line with 1 1 . . . • c l and 1 1 . . . •b6
White a clear edge, while 1 9...b 6 has 32 l:l.xd 5!? exd5 33 iLxd5 'iia 7!
been suggested by Watson, though I Other moves lose. Nijboer gives the
don't trust that at all after 20 'ii'd3 following lines:
(threatening b2-b4) as 20 ... lbcxd4 fails a) 33 ... i.e8 34 e6 l::t e7 35 :f3!, in
to 2 1 lbxd4 i.xa4 22 lbxf5 :Xf5 23 tending :xf8+ and 'ii'h8 mate.
i.xa4 'ii'xa4 24 :xe6 :fl 25 :xe7 b) 33 ...i.f5 34 g4.
:xe7 26 'ii'xdS+. c) 33 ...l::td8 34 i.xf7+ 'it>xf7 35 l:r.d3.
20 hx93 'iic7 21 l:.c 1 ! 34 iLxf7+ �xf7 35 'iid6 l:.e8?
35 ...'it>g8 isn't mentioned by Ni
jboer and seems like a tougher de
fence. After 36 e6 i.e8 37 :£3 h6 38
Game 1 7
Majer-Ruzicka
Correspondence 1993
1 e4 e6 2 d4 d5 3 tlld 2 tllf6 4 e5
43
The Fre n c h Ta rra s c h
44
3. . . li:J f6 : Main Line with 1 1 . . . "ii c l a n d 1 1 . . . "ii b 6
45
The Fre n c h Ta rra s c h
the bishop pair, but it soon becomes 31 'it>xg4 'ii'fS+ 32 'it>h4 :h3 is mate.
obvious that the most important as A convincing display from Martin.
pect of the position is his shaky king.
20 l:tae 1 tllg4+ 21 �g1 'ii'd 6 22 g3 Game 19
l:l.f3 23 �g2 Muratov-Bata
23 .tf4 :xf4! 24 gxf4 'ii'xf4 25 'it>g2 Correspondence 1967
:f8 is the end.
23 . . .l:Iaf8 24 'ii'b4 'ii'f6 25 iLe2 1 e4 e6 2 d4 d5 3 tlld 2 tll f6 4 e5
tllfd7 5 il.d3 c5 6 c3 tllc6 7 tlle2
'ii'b6 8 tllf3 cxd4 9 cxd4 f6 1 0 exf6
tllxf6 1 1 0-0 il.d6 1 2 tllf4
46
3. . . tll f6: Main Line with 1 1 . . . 1ic7 and 1 1 . . . 1i b 6
47
The Fre n c h Ta rra s c h
1 5 . . . 96
Two other moves have also been
witnessed from this position: 22 . . .tllf7 23 l:.e2 ilb5 24 l:l.e3 'ii'd6
a) 15 ...lbe4 16 g3 lUS!? 17 lbh4 :£7 25 l:l.ce 1 tllda 26 l:l.e5 ilea 27 h5
18 lbf3 nfs 19 lbh4 :fl 20 lbf3 and 'ii'd7 2a hxg6 hxg6 29 l:.g5 l:l.xf4 30
the players agreed a draw in Wahls gxf4 .Uf6 31 'ii'e3 ilf7 32 �g2 tll c6
Kindermann , Dudweiler 1996. 17 33 l:l.h 1 ilea 34 l:.h3 'ii' g7 35 f5 ilf7
l:te l!?, hassling the knight on e4, looks 36 fxg6 1 -0
48
3 . . . tll f6: Main Line with 1 1 . . . 1i c7 a n d 1 1 . . . 1i b 6
Summary
The plan of i.g5-h4-g3, in conjunction with leaving the knight on e2, continues
to be the most dangerous plan against 1 1 ...'ii'c 7. Black has to play extremely care
fully to avoid drifting into a lifeless position in this line. On the other hand it is
still unclear how White should proceed against 1 1 ...'ii'b 6. Many different ways
have been tried but Black seems to have enough resources in each variation.
1 1 . .'ii'c7
.
1 1...'ii'b6
12 ltJc3 0-0 (D)
13 .tgS
13 ....td7 Game16 -
13 i.e3 Game18
-
12 ltJf4 Game19-
12 b3 Game20
-
12 il.g5 (D)
12 ltJc3 Game14
-
12 g3 Game15
-
1 3 ... es
13 ... ltJhS Game12
-
14 dxe5 - Game11
49
CHAPTER THREE I
3 '2Jf6: Early Deviations
. . .
50
3 . . . tll f6 : E arly De viati o n s fro m t h e Main L ine
reach a position similar to the one dis this strategy proves to be mightily
cussed in Chapter 2. Theoretically effective.
White maintains an infinitesimal edge 1 4 a3
and Black's position is solid, but 14 'ii'h7+ Wf7 is just a waste of time.
slightly depressing. Note that if Black The queen would soon have to beat a
wishes to play with ... .td6, then hasty retreat after ... lbf8.
10 ... h6 is the most accurate move or 1 4 . . .tllfB 1 5 b4 il.d7 1 6 iLb2 b5 1 7
der, preventing White's option in the iLc3 ilea 1 8 'ii'e3 il.96
next note. The plan has been completed and
Black stands well. Now White chooses
not to exchange, but this merely leaves
Black with two good bishops against
two bad ones.
1 9 il.b3 tlld7 20 l:tae1 .Ue8 2 1 tllg3
iLf4 22 'ii'e2 a6 23 tlle5 tlldxe5 24
dxe5 'ii'g5 25 iLb2 %lad8
1 1 0-0
Here 1 1 .tgs 'ii'f7 12 0-0 0-0 13
.th4, threatening lbgS and also intend
ing the positionally desirable .tg3,
gives White a traditional plus. Hence
the growing popularity of the move
10... h6.
1 1 . . . h6 1 2 iLc2 An uncomplicated plan. The d
The alternative 12 .th 1 is the sub pawn will roll up the board. White's
ject of Game 22. next move is a sure sign that things are
1 2 ... 0-0 1 3 'ii'd3 slipping.
White's play looks very natural, but 26 tllh 1 iLh5 27 f3 d4 28 tllt2 'ii'h4
in fact it founders on an original de 29 tllh3
fensive manoeuvre. 29 g3 .txg3! 30 hxg3 'ii'xg3+ 3 1 Whl
1 3 . . . l:.dS ! .txf3+ is the end.
Black's fortifying strategy includes 29 . . .iLe3+ 30 �h 1 tllxe5 3 1 'ii'c 2
...lbf8, covering the weak light squares iLxf3 32 gxf3 'ii'xh3 33 'ii'e4 tll f7 34
around the king. This will be followed f4 tlld6 35 'ii'e 5 tll c4 36 iLxc4 bxc4
by ... i.d7-e8-g6! (or hS), after which 37 l:.g 1 il.xg 1 38 %lxg 1 'ii'f3+ 39 l:.g2
Black's pieces will co-ordinate per 'ii'd 1 + 40 l:.g 1 'ii'f3+ 41 l:.g2 l:.e7 42
fectly. Simple stuff, but in this game 'ii'f6
51
The Fre n c h Tarra s c h
52
3 . . . ltJ f6 : Ea rly De viations fro m t h e Main L ine
53
Th e Fre n c h Ta rra s c h
21 . . . il.g4!! 0-1
After 22 l:txg4 Black mates with
22 ...'ii'f l+ 23 :gl lbg3+! 24 hxg3 'ii'h3 .
Game 24
Lputian-Agzamov
USSR Championship 1985
1 d4 e6 2 e4 d5 3 tlld2 tllf6 4 e5
tllfd7 5 il.d3 c5 6 c3 tllc6 7 tlle2
1 4 . . . e5 1 5 tllf3 tllxf3+ 1 6 gxf3 tllh5 cxd4 8 cxd4 tll b6
1 7 il.xg6+
Drastic measures are required, oth
erwise Black will consolidate with
...'ii'f6.
1 7 . . . �xg6 1 8 �h1 'ii'h 4!
Black does best to give back the
piece. 1 8 ... lbf6 allows 19 :gl+ 'it>f7 20
:Xg7+ 'it>e6 21 .tgS .te7 22 l:.xe7+!
'ii'xe7 23 l:.cl!
1 9 'ii'xf8 �h7 ?!
It is better to play 19 ...'ii'h 3 20 :gl+
'it>h7 21 'ii'a3 .tfs 22 :g2 d4, when
Black had good compensation in 9 0-0
Doric-D'Amore, Formia 1995. After Here White can also take immediate
the text move, 19 ... 'iti>h7, White should steps on the queenside with 9 a3 aS 10
play 20 'ii'f7! b3. At first sight these moves seem a
20 l:l.g 1 ? ! 'ii' xf2 2 1 'ii'f 7?? little strange, but White is taking pro
One move too late. This allows a phylactic measures. Black's... a5-a4 will
very pleasing finish. now be answered by b3-b4, establish-
54
3 . . . liJ f6 : Ea rly De viations fro m t h e Main Line
55
Th e Fre n c h Ta rra s c h
Game 26
Emms-Strater
Hastings 1996
56
3 . . . tll f6 : Early D e viations fro m t h e M a in L ine
57
The Fre n c h Tarra s c h
1 3 . . . 'ii'c5
The most popular move, although
White has tended to score well against
3 1 'ii'x f5 ! ! exf5 32 iLxe5 'ii'd7 33 this. Major alternatives include (in de
l:th6 f6 34 iLxf6 'ifa4 35 .Ue 7 1 -0 creasing order of importance):
Black can only prevent mate with a) 13 . ..ltJcS!? 14 i.d2 'ii'a4 15 b3
the ludicrous 35 ...'ii'a l+ 36 i.xal 'ii'd7 16 i.e2 i.e7 17 i.e3 b6 18 b4
'it>xe7, after which White is too much 'ii'a4 19 'ii'h2 ltJe4 20 a3 i.a6 21 bs
material up. i.b7 22 .Ufc l l:r.c8 23 ltJd4 0-0 24 f3
.Uxcl + 25 .Uxcl i.xa3 26 'ii'a2 'ii'b4
with great complications in Hellers
Game 27 Ludvigsen, Nordic Team Champion
Kolev-Herraiz ship 1985.
Linares Open 1996 b) 13 ...i.e7 14 i.d2 'ii'g4 15 l:tac l 0-0
16 i.xh7+ 'it>h8 17 i.d3 ltJcS 18 h3
1 e4 e6 2 d4 d5 3 tlld 2 tllf6 4 e5 'ii'hS 19 i.gS with a distinct advantage
tlltd7 5 il.d3 c5 6 c3 tll c 6 7 tllgf3 for White in Emms-Tibensky, Douai
Introduced by Vik.tor Korchnoi in Open 1990.
the 1960s, and later advocated by John c) 13 ... h6?! 14 i.d2 'ii'b6 15 :ac t
Nunn, this leads to a speculative offer i.e7 16 'ii'a4! is given as clearly better
of a pawn. for White by ECO, as 16 ... 0-0 allows
7 . . . 'ii'b6 8 0-0 cxd4 17 :Xc8 .Uaxc8 18 'ii'xd7.
8 ... g6 is discussed in Game 28. 1 4 'ife2 iLe7 1 5 iLe3 'ifa5 1 6 .Uac 1
9 cxd4 tll xd4 1 0 tll xd4 'ii'xd4 1 1 tllf3 0-0 1 7 l:Z.c3 l:.e8
'ii'b6 1 2 'ifa4 'ii'b4 Th alternative 17 .. .fS 1 8 exf6 ltJxf6
White's idea was to swing the queen 19 i.d4 offers White chances of a di
to the kingside with, for example, rect attack. Emms-Giddins, Gausdal
12 ... i.e7 13 'ii' g4. 12 ...'ii'b 4 prevents 1993, concluded 19 ... i.d7 20 ltJgS h6
this. 2 1 ltJxe6 l:tfe8 22 i.f5 'ii'a6 (or
1 3 'ii'c 2 22 ... i.bS 23 'ii'd2 i.xfl 24 ltJxg7! 'it>xg7
58
3 . . . li:J f6 : E a rly De viations from t h e M a in L ine
2 5 .Ug3+) 2 3 'ii'xa6 bxa6 2 4 lbxg7 i.xf5 White. Note that 21...lbxe5 allows
25 lbxf5 lbe4 26 l:tc6 1-0. After mate after 22 :h3 lbg6 23 'ii'h7+ 'it>f8
17 ... .Ue8 Black doesn't weaken the 24 'ii'h8+ lbxh8 25 l:.xh8.
kingside, but does allow an effective
version of the Greek gift.
59
The F re n c h Ta rra s c h
Game 29
Emms-S.Clarke
British Championship 1991
1 e4 e6 2 d4 d5 3 tlld 2 tll f6 4 e5
tll fd7 5 tllgf3 c5 6 c4?!
An interesting possibility, but not
one to cause Black any great deal of
concern. I tried this line out a few
The opening battle has been well times, but this game was my last expe
and truly won by Black. The rest of rience with it.
60
3 . . . liJ f6 : Ea rly D e vi·ations from the Main Line
:..:....:.
___
: 2 lt:Jc5 31 t5
9
3 1 ...gxf5 32
� :xf5 is strong.
gxf d
� � ��: �
. r•. ·· i
-
%-, 1 � � �.,,
� J�.
-,
6 , , .lt:Jc6 7 cxd5 exd 5 8 i.d3 cxd4 9
•. .
t.•
�*if'�� �, -
·� � ··1
!ff{���-' "�,.ft
0-0 lt:Jc5 1 0 lilb3 � 4 1 1 lt:Jxc5
9
.txc5 1 2 a 3 i.b6 1£ 3 :e 1 0 -0 1 4
��� �� !i. •
B
JU:,� � ;
i.c2 i.h5!
Planning . .. 1'.g6 . Apart from the
. - :
;�
. doub led' White has no
. is
fact that lt .
real compensat n for the pawn. 37 i.h4!
��
1 5 i.t4 :ea "illd 3 .'11.g6 1 7 "ill� 2 Turning the tabl White threatens
:c8 18 i.xg6 hxg6 1 9 :ad 1 'Wd7 O 38 'ir'h7+ @f8 39 'iV + 'ir'g8 40 1'.e7+.
h3 a5 2 1 g4 a4 22 'Wd3 nae 23 :e2 37 . . .:ta 38 'Wh7+ 'it>t8 39 i.xt6 1 -0
:as 24 :de 1
Game JO
Seu I-Zach
Biel 1997
1 e4 e 6 2 d4 d5 3 '"UC
'""' 3 lt:Jt6 4 e5
tbe4 5 lt:Jxe4 dxe4
24 , , .lt:Jd8'
:
Returning the extra pawn in ex-
change for a · �ood knight versus bad
.
bishop' scenario. White also has to
worry about Black's other d-pawn.
25 lt:Jxd4 i.x 4 26 "illxd4 lile6 27
�
°Wd2 :b 5 28 i.g 3 d4 29 t4 :b3 30 6 .>11.e3
�
�
-_
______
66 11
_
-------
The F re n c h Tarra s c h
Note that although White played 3 two bishops and kingside pawn pha
tt:Jc3 in this game, T arrasch players lanx offer reasonable compensation.
would reach the same position with 3 1 5 . . . .ta3?
tt:Jd2. The other main alternative here Black just loses valuable time with
is 6 i.c4, intending to ignore rather this move. Perhaps he should try
than surround the e4-pawn. However, 15 ...Itd7 16 @b l i.b4 17 lDxe4 :hd8
the pawn does have some nuisance 18 c3 i.d5 19 a3 i.e7, with some pos
value. After 6 ...a6 7 a4 b6 8 tt:Jh3 i.b7 sible play on the light squares around
9 tt:Jf4 tt:Jc6 10 i.e3 tt:Je7 1 1 0-0 g6 1 2 the white king. As the game goes,
'ir'e2 tt:Jf5 13 IHdl i.h6! Black retained Black gets no compensation for the
counter-chances in Aseev-Lputian, pawn.
Russia 1984. Indeed Black was able to 1 6 'We3 .te7 1 7 'it>b 1 l::.d7 1 a lt:Jxe4
seize the advantage after 14 a5 0-0 15 l::.hda 1 9 l::.d3 'it>ba 20 l::.hd 1
axb6 cxb6 16 i.xa6?! i.xf4! 17 i.xb7
tt:Jxd4 1 8 'ir'c4 l:hal 19 l:hal i.xe5 20
i.xe4 'ir'h4. After 6 i.e3 Black has to
play very actively, else the e-pawn will
simply disappear for nothing.
6 . . . c5 7 dxc5 lt:Jd7 a 'Wg4 lt:Jxc5 9
.tb5+ lt:Jd7 1 0 lt:Je2 'Wa5+ 1 1 lt:Jc3 a6
1 2 .txd7+ .txd7
62
3 . . . ll:i f6: Ea rly De via tions fro m the Main Line
1 1 iL.f1 !
No more chances! Now White's
pawn wedge on e5 guarantees a nice
space advantage.
1 1 . . .'irb6 1 2 iL.c3 iL.d7 1 3 iL.d4 �c 7
1 4 l:t.e3 a5 1 5 :tc1 b6 1 6 c3 a4 1 7
8 lt:Jf3 lt:Jxc5 9 0-0 iL.e7 g3 iL.c6 1 8 h4 :aea 1 9 iL.h3 ll:le4 20
On this or the next move Black ll:ld2 f5 21 exf6 iL.xf6 22 iL.xf6 l:t.xf6
should probably capture on d3. One 23 lt:Jxe4 dxe4 24 �e 1
recent example of such play was Black has eliminated the e5-pawn,
G.Mohr-Sulava, Croatian Cup 1997, but is left saddled with a chronically
which continued 9 ...CfJxd3 10 cxd3 weak pawn. As soon as this is rounded
i.d7 1 1 i.g5 'iia5 1 2 a3 Itc8 13 Ite l up, Black sees no reason to continue
.l::!.xc l 1 4 'iix cl h6 1 5 b4 'ir'a6 16 i.e3 the game.
ii.el 17 'ir'c7 i.d8 18 'iic5 b6 19 'iid6 24 . . . b5 25 :.d 1 iL.d5 26 a3 l:t.ef8 27
i.e7 20 'iib 8+ i.d8 2 1 'ir'd6 i.e7 and l:t.d2 l:t.f3 28 iL.g2 l:t.xe3 29 �xe3 �e5
the players agreed a draw. 30 l:t.d4 1 -0
63
The F re n c h Ta rra s c h
Summary
The 3 ...lbf6 Tarrasch offers plenty of offbeat variations for both sides that are
worth an outing. Of course, due preparation is advisable before venturing out
with a new line.
1 e4 e6 2 d4 d5 3 lt:Jd2 lt:Jf6 4 e5
4...lt:Jfd7
4...lbe4
5 lbxe4 - Game30; 5 .id3 - Game31
5 .td3
5 lbgf3 cS
6 c3 lbc6 7 i.d3 - see Games 27 and 28 (by transposition)
6 c4 - Game29
5 ... c5 6 c3 ttJc6
6 ...b6 - Game26
7 tt:Je2
7 lbgf3 'iib 6 8 0-0 (D)
8 ... cxd4 Game27
-
8 ... g6 - Game28
7 . . . cxd4
7 ...
- 'iib6 8 lbf3 cxd4 9 cxd4 f6 (D)
10 exf6 lbxf6 1 1 0-0 .id6 - Games 16-20
10 lbf4 - Game25
8 cxd4 f6 (DJ
8 ...lbb6 Game24 -
9 exf6
9 lbf4 - Game23
9 ...'Wxf6 1 0 lt:Jf3 h6 1 1 0-0 .td6 1 2 .tc2
12 .ib l Game22
-
1 2 . . . 0-0 Game21
-
8 0-0 9 . f6
. . 8 . .f6
.
64
CHAPTER FOUR I
3 tiJf6: The Big
. . .
3 ... lbf6. He bolsters the eS-pawn and queenside play with ... b7-b5-b4.
plans to develop his pieces behind an
impressive centre, before slowly Game 32
squashing Black on the kingside. Emms-Kosten
That's the theory, and it happens quite British Championship 1985
a lot in practice too, but of course
Black does have some moves too. The 1 e4 e6 2 d4 d5 3 lt:Jd2 lt:Jf6 4 e5
downside to 5 f4 (compared to 5 .id3) lt:Jfd7 5 f4 c5 6 c3 lt:Jc6 7 lt:Jdf3 cxd4
is that it doesn't contribute towards 8 cxd4 ._b6 9 g3 .tb4+ 10 �2 g5!?
White's development. Consequently,
Black can obtain quicker and more
dangerous counterplay against the d+
pawn. Another facet of this line is that
White is often forced to go on a little
walk with his king, which is not to
everyone's taste. Normally it would
be completely safe behind that huge
phalanx of pawns, but Games 32-34
show that Black does have some in
genious ways of causing a few prob
lems. Game 35 sees a relatively new
concept for White, the lavish 8 h4, This move turned a fe w heads when
which puts paid to any ... g7-g5 breaks. it was first unleashed on the touma-
65
The Fre n c h Tarra s c h
66
3 . . . ll:i f6 : The B ig Pawn Fro n t with 5 f4
67
The Fre n c h Ta rra s c h
1 3 . . . lt:Jc4!
Naturally 13 ... lbc6 is also possible,
but this seems to be the most reliable
way of equalising. After the exchange 1 0 .th3 fxe5 1 1 fxe5 .tb4+ 1 2 'itf1
on c4 Black can lay claim to many 0-0 1 3 'itg2 lt:Jdxe5 ! ?
important light squares. Black strikes before his opponent
1 4 .txc4 dxc4 1 5 lt:Jf3 i.d6 1 6 'Wc2 can consolidate his position.
'Wc6 1 7 :he 1 0-0 1 4 dxe5 lt:Jxe5
There also seems nothing wrong
with the immediate 17 ... bS, planning
1 8 ... 1'.b7.
1 8 .tf4 b5 1 9 .txd6 .tb7! 20 'We4
'Wxe4 2 1 :xe4 .txe4 � - �
After 2 1.. .1'.xe4 22 1'.xf8 @xf8 it is
about equal, although there is still
plenty of fight left in the position.
Game 34
Mohrlok-Schrancz
Correspondence 1989
1 5 'We2!
1 e4 e6 2 d4 d5 3 lt:Jd2 lt:Jf6 4 e5 This has been shown to be the only
lt:Jfd7 5 c3 c5 6 f4 lt:Jc6 7 lt:Jdf3 cxd4 real major test of Black's sacrifice.
8 cxd4 'Wb6 9 g3 f6 Against other lines Black seems to be
Before the discovery of 10 ... gS, this holding his own at least.
was Black's most risky and challeng a) 15 't\Vb3 1'.cS 16 't\Vxb6 1'.xb6 17
ing try against 5 f4. Black is prepared 1'.f4 lbd3 18 1'.d6 19 IUl eS! 20 1'.xeS
to sacrifice a piece in order to destroy (20 lbxeS?? lbe l+! 21 Itxe l Itf2 is
White's central pawn cover. Note that mate) 20 ... 1'.xh3+ 21 lbxh3 Ite8! 22
the move order is quite interchange- Ite 1 Itfe7! and Black regained the piece
68
3 . . . tli f6 : Th e B ig P a wn Fro n t with 5 f4
20 . . . :ctS
Following a suggestion of Knaak's.
20 ... Ite8 21 'ir'b5! was also good for 1 1 .te31 .te7 1 2 'Wd21
White in Ermenkov-Knaak, Bulgaria This is more accurate than 12 Itbl,
East Germany 1983. Instead, damage e.g. 12 ...aS 1 3 lbe2 0-0 14 lbg3 fxe5 15
limitation can be achieved with fxe5 and now Black can try the typical
20 ... Itxcl 2 1 �hcl g5 22 Itc7! (22 'ir'g4 sacrifice 15 ...lbdxe5! 16 dxe5 d4 17
.l::!.xe5 23 Itc8+ 1'.f8! 24 'ir'xe6+ Itxe6 25 1'.d2 lbxe5 .
i.xg5 @f7 is equal) 22 ... 1'.d6 23 I:.xb7 1 2 . . . 0-0 1 3 :d 1 a5!
i.xe5 24 1'.d2!, which Mohrlok as Brutal attempts do not work here.
sesses as only slightly better for White. After 13 ... fxe5 14 fxe5 lbdxe5 15 dxe5
2 1 'We3! d4 White holds on to the e5-pawn
69
The Fre n c h Ta rra s c h
Game 36
Adorjan-J.Watson
New York 1981
We now enter a heavy manoeuvring
stage of the game, where Black's 1 e4 e6 2 d4 d5 3 ll:ld 2 lt:Jf6 4 e5
chances are not worse. ll:lfd7 5 c3 c5 6 f4 lt:Jc6 7 lt:Jdf3 'Was
1 7 h5 ll:lb6 1 8 'iitf 2 ll:lb4 1 9 .tb 1
ll:lc4 20 'Wc3 .td7 2 1 a3 ll:lc6 22
l::.d g1 b5 23 g4 b4 24 axb4 a3! 25
gxf5 exf5 26 .ta2 .te6 27 ll:lg5 'Wd7
28 lt:Jxe6 'Wxe6 29 .txc4 .txb4 30
'Wc2 dxc4 3 1 bxa3 .txa3 32 .td2
l::.fd8 33 .tc3 .te7 34 °Wd2 'Wd5 35
l::. h 3 l::. a3 36 'Wc2 'We6 37 :1h2 l::.da8
38 I:f. b 1 l::.b3 39 l::.d 1 'Wd5 40 'i'xf5
.l::t a 2 41 'Wh3 ll:lb4 42 'itg 1 lt:Jd3 43
'Wea+ .tta 44 e6 :as?
44 ..Jhe2! would have capped off an
excellent positional performance from Adding pressure to the centre by
the German grandmaster. After 45 pinning the c-pawn. This move was
I:.xe2 lbxf4 46 e7 lbxe2+ 47 @f2 'iVf7+ popular in the seventies, but is hardly
Black is winning, e.g. 48 @xe2 't\Vxe7 + ever seen nowadays. Another quieter
49 @d2 't\Vg5+ 50 @c2 'iYg2+ 5 1 Itd2 move is the blocking 7 .. .f5. Ye Jiang
't\Ve4+ 52 @dl Itbl mate. 44 ... Ita8 al chuan-Short, Lucerne 1989, continued
lows White to scrape half a point. 8 i.d3 cxd4 9 cxd4 Ji.el 10 lt:Je2 lbb6
70
3 . . . lU f6 : The Big Pawn Fro n t wit h 5 f4
1 0 lt:Jd4
This move, a big novelty at the
time, has put the whole line to rest.
Previously 10 a3 had been the main
move, but Black gets sufficient play Black has managed to regain some
with the pawn sacrifice 10 ...b3! 1 1 of his lost material, but now the four
'ir'xb3 1'.xcS. White minor pieces run rampant, cre
1 0 . . .'Wxc5 ating deadly threats around the black
Or 10 ... 1'.b7 1 1 a3! bxc3 12 b4 'ir'd8 king.
13 ltJgf3, when the c3-pawn will soon 24 . . .:thcS 25 ll:lc6+ l:txc6 26 .txc6
die and Black will be swamped by the :ca 27 .tg5+ �e6 28 lt:Jd4+ �xe5
mass of queenside pawns. 29 .txd7 :tc 1 30 ll:lf3+ '1i;d6 3 1 .tf5
1 1 .,a4 bxc3 .tc5+ 1 -0
71
The Fre n c h Tarra s c h
Summary
It seems that Black has several ways of creating enough confusion against
White's impressive looking centre. As of yet, no real answer has been found to
10 ... gS, and this is probably the single biggest reason for the decline in popularity
of 5 f4.
7 1i'b6
... (D)
7 ...'ir'aS - Game 36
8 g3
8 h4 - Game 35
8 . . . cxd4 9 cxd4 .tb4+ (D)
9 .. .f6Game 34
-
10 �2 g5 11 .te3 (D)
1 1 fxgS Game 33 -
1 1 . . . f6 Game 32
-
7 "flb6
...
9 ..tb4+
. . . 1 1 ..te3
72
CHAPTER FIVE I
3 cs 4 exd5 'i'xd5 :
M a i n Line with 10 ttJxd4
. . .
ten White's attempts at a direct attack leads to quieter positions where Black
crash disastrously against this solid has less chance of being rolled over,
defensive wall of pawns. Black also but White often secures the long-term
tries to gain time by developing with advantage of the bishop pair. That
tempo (e.g. ...'if cl, attacking c4, and said, there is also plenty of scope here
... 1'.d6, hitting the h2-pawn) . From for both sides to liven up the proceed
humble beginnings this variation is mgs.
now firmly established as a favourite
defence of some of the world's leading Game 37
players. Bowden-Levitt
From the starting position of the England 1997
main line Black has two main options,
10 . a6 and 1 0 .. 1'.d7. Both moves pre-
. . . 1 e4 e6 2 d4 d5 3 lbd2 c5 4 ll:lgf3
pare a later ...'i'c7 and . . . 1'.d6, but cxd4 5 exd5 .-xd5 6 .tc4 'Wd6 7 0-0
10 ... a6 (Games 37-42) has generally ll:lt6 8 ll:lb3 ll:lc6 9 ll:lbxd4 ll:lxd4 1 0
73
The F re n c h Tarra s c h
1 9 .tc1
Following Michael Adams's recipe
This concept is well known here. (see below) . This move, threatening
White readily sacrifices material for an 't\Vg4+, has the advantage of not block
attack against the black king. Despite ing the e-file, which can be vital in
the fact that it has been tested in nu many variations. However, Jonathan
merous grandmaster games, no firm Levitt had prepared an extremely sig
conclusions have yet been reached. nificant novelty. The other critical
The quiet 13 h3 causes Black fewer line is 19 1'.e3, as played by Sergei Ru
problems, as we shall see in Game 39. blevsky (Rublevsky has worked with
1 3 . . ..txh2+ 14 'ith1 0-0 1 5 l2Jxg 7! Kasparov, so on that evidence alone
:dB! we must take this move seriously!):
Apparently this defensive resource a) Levitt's 19 ...1'.f4 is still possible,
was Kasparov's discovery. Earlier this but not quite so effective, as after 20 g4
whole line had been deemed dubious ltJg3+ 21 fxg3 1'.xe3 22 'ifxe3 White's
for Black on account of 15 ...@xg7 16 pieces are better co-ordinated than in
't\Vd4 (planning 'ifh4 and i.h6+) 16 ...eS the main game. In particular, the
17 'iVh4 ltJg4 18 f3 'ifd8 19 i.gS f6 20 white queen patrols the dark squares
i..c l ! and Black is in trouble. Emms around the black king. Nevertheless,
W.Lowe, Halifax Open 199 1, contin Black still seems to be just about okay.
ued 20 ...'ifd4 2 1 fxg4 1'.f4 22 c3 'iVcS For example, 22 ...h6 23 1'.c2+ @g7 24
23 1'.xf4 exf4 24 Itadl and White had a gS hs 25 .l::!.adl b6 26 'iVf4 i.b7+ 27
winning advantage. @h2 'iVxf4 28 gxf4 .tf3 29 Itd3 1'.g4 30
1 6 °Wf3 'itxg7 1 7 .th6+ 'itg6 @g3 f 6 31 Itee3 1h-1h was the conclu-
74
3 . . . c S 4 e x d S 'ik xd S : Main Line with 1 0 Ei:J x d 4
75
T h e Frenc h Ta rra s c h
22 :xd5
Game 38
Adams-Dreev
Wijk aan Zee 1996
1 e4 e6 2 d4 d 5 3 lt:Jd2 c5 4 exd5
'Wxd5 5 lt:Jgf3 cxd4 6 .tc4 'Wd6 7 0-0
ll:lf6 8 ll:lb3 ll:lc6 9 lt:Jbxd4 lt:Jxd4 1 0
lt:Jxd4 a 6 1 1 :e 1 'Wc7 1 2 .tb3 .td6
1 3 ll:lf5!? .txh2+ 1 4 'ith 1 0-0 1 5
ll:lxg7! :dB! 1 6 °Wf3 'itxg7 1 7 .th6+
'itg6 1 8 c3 ll:ld5 1 9 :ad1 !
76
3 . . . c S 4 e x d S 'ik x d S : M a in Line with 1 0 t'iJ x d 4
cxd4 5 exd5 'Wxd5 6 ..tc4 'Wd6 7 0-0 lems, as 20 g3 can be simply answered
.!Lif6 8 ll:lb3 ll:lc6 9 .l::.e 1 a6 1 0 ll:lbxd4 by 20 ...1'.xg3 with a mating attack.
ll:lxd4 1 1 ll:lxd4 'We 7 1 2 ..tb3 ..td6 1 3 20 f3 J:r.ag8 2 1 g4 'Wc5 22 'itg2 ..tc7
h3 0-0 1 4 ..tg5 b5! 23 J:r.e2 'Wd6 24 .l::.h 1 a5!
Black temporarily concentrates on
the queenside, with the aim of de
stroying the support for the knight on
d4. Now 25 lDxbS runs into 25 ... 'ir'g3+
26 @f1 1'.d8!, trapping the queen.
25 ..tc2 b4 26 ..td3 a4 27 h4
Understandably White is desperate
for counterplay, but this move seri
ously weakens the pawn on g4.
27 . . . ..td8 28 'We5 'Wxe5 29 .l::.x e5
l:bg4+ 30 'itf1 .l::.f4
Game 40
Adams-Speelman
Black's dangerous initiative is worth England 1991
at least a pawn. Note that White's ear
lier h2-h3 has presented fresh prob- 1 e4 e6 2 d4 d5 3 ll:ld2 c5 4 exd5
77
The Fre n c h Tarra s c h
78
3 . . . c s 4 e x d S • x d S : M a in Line with 1 0 liJ x d 4
The last chance was 26 ...'ir'b5. 1 1...'iVcl can transpose to the main
27 'Wc4 ll:ld5 28 .tb6 ll:lxb6 29 'Wc7+ lines after 12 Ite 1. White also has the
�as 30 axb6 independent try 12 'ir'f3, but after
Black's back-rank weaknesses are 12 ...1'.d6 13 h3 0-0 14 1'.g5 lbd7 15 c3
absolutely fatal. b5 16 Itadl lbc5! 17 1'.c2 1'.b7 18 'ifh5
1'.e4 Black exchanged the dangerous
bishop and had no problems in
Hellers-Hiibner, Wijk aan Zee 1986.
12 c 3 'Wc7
Also possible is 12 ... 0-0-0 13 Ite l
'ir'c7 14 'ir'e2 i.d6 15 h3, transposing
to the previous game.
1 3 .tg5 h5!?
Black has three alternatives here:
a) 13 ...lbe4 14 i.h4 'ir'f4 15 g3 'ir'h6
16 'ir'e2 lbd6 and now Glek gives 17
'ir'e5, intending 17 .. .f6? 18 lbxe6!! fxe5
30 . . .'ii'd 7 3 1 lbe5 nta 32 b5 1 -0 19 lbc7 mate.
After 32 ...'ifxc7 33 bxc7 Ik8 34 b6 b) 13 ...1'.d6 14 1'.xf6 gxf6 15 'ifh5
and White can win at leisure. 'ir'c5 16 'ir'f3 'ii'e 5 17 g3 0-0-0 18 .l::!.fe l
'ir'g5 1 9 i.c4!, with the idea 1'.f1-g2,
Game 41 when White had an edge in Kudrin
Svidler-Glek Remlinger, Philadelphia 1992.
Haifa 1996 c) 13 . . 0-0-0 14 1'.xf6 gxf6 15 'ir'h5
.
79
The F re nc h Tarra s c h
Game 42
G. Timoshenko-Daniel ian
2 2 . . .ll:lxf2+ 2 3 'it h 2 'Wd8 2 4 :es Cappelle la Grande 1994
.tb6 25 'Wd6
Threatening 'ir'b8 mate. 1 e4 e6 2 d4 d5 3 ll:ld2 c5 4 exd5
25 . . . ll:lxd 1 'Wxd5 5 ll:lgf3 cxd4 6 .tc4 'Wd6 7 0-0
25 ...i.c7 leads to an amazing posi lt:Jf6 8 ll:lb3 ll:lc6 9 ll:lbxd4 ll:lxd4 1 0
tion after 26 Ik5 1'.c6 27 l:hc6!! bxc6 lt:Jxd4 a6 1 1 b 3 'Wc7 1 2 .tb2 .td6 1 3
28 'ifxc6 lbxd1 29 lbd4!! (Har-Zvi). ll:lf3 b6! ?
80
3 . . . c S 4 e x d S 'ii x d S : Main Line with 1 0 tlJ x d 4
13 ...b6 permits this without allowing 'ii'x h2+ 31 <it>e3 l::t e5+ 32 llle4 llld5+
White to open up the position with 0- 1
a2-a4. 33 @d4 1i'b2+ 34 @c4 .l::!.xe4+ 35
1 4 l::te 1 il.b7 1 5 il.d3 dxe4 1i'c3 is mate.
15 i.xe6 fxe6 16 .l::!.xe6+ looks entic A pleasing game from the Arme
ing, but after the simple 16 ...@d7 I nian; 13 ... b6 looks like a very useful
can't find a decent way to continue move.
the attack.
1 5 . . . llld 5! Game 43
Offering a long-term pawn sacrifice, Emms-Bibby
since after 16 i.xg7 .l::!. g8 17 i.b2 lt:Jf4 British Championship 1990
18 g3 lt:Jxd3 19 1i'xd3 0-0-0 Black has
play down the half-open g-file and a 1 e4 e6 2 d4 d5 3 llld2 c5 4 lll gf3
dominating bishop on the long diago cxd4 5 exd5 'ii'x d5 6 iLc4 'ii'd 6 7 0-0
nal. All in all, this amounts to excel lllf6 8 lllb3 lll c6 9 lllbxd4 lllxd4 1 0
lent compensation for the pawn. lllxd4 il.d7
1 6 a4 0-0-0 1 7 g3 <it>b8 1 8 il.xg7?
White's position was already noth
ing to write home about, so I imagine
Timoshenko's philosophy was that he
may as well have a pawn for his trou
bles. Unfortunately, this runs into
some real danger.
1 8 . . . l::th g8 1 9 iLe5 iLxe5 20 lllxe5 f6
21 lllc4 lllc 3 22 'ii'h 5 'ii'c6
81
Th e Fre n c h Tarra s c h
82
3 . . . c S 4 e x d S 'ik x d S : Main L ine with 1 0 t:fJxd4
83
The Fre n c h Ta rra s c h
21 .. .l::txh2 ! 22 iLxh2
22 @xh2 'ifh6+ 23 @gl i.xg2! wins
beautifully after 24 @xg2 1i'h3+ 2S
@gl 'ifxg3+ 26 @h 1 1i'h3+ 27 @gl g3.
Now White bails out for a draw.
1 5 h3 22 . . . g3 23 il.xg3 'i'xg3 24 "iif 1 "iif4
The other main try is lS l:.adl. Af 25 :td 1 il.d6 26 l::txd6 1h -1h
ter 1S ... i.d6 16 h3 i.c6 17 l:!fe l i.cS The position is completely level.
84
3 . . . c S 4 e x d S 'ik x d S : Main Line with 1 0 t:fJxd4
Summary
I don't believe for a moment that we've heard the last of the 1 1 %:!.e l iic7 12 ..5lb3
..5ld6 13 lt:JfS!? argument, the assessment of which seems to change from day to
day. One thing is clear: both players have to know their stuff well before enter
ing into the labyrinth of complications. White seems able to keep a nagging edge
against 10 ... ..td7 with the simple 1 1 c3 and 12 'iie 2.
1 0 . . . a6
10 ... ..td7 (D)
1 1 c3 iic7 12 iie2
12 ...0-0-0 Game 43
-
1 1 b3 Game 45 -
1 1 :e 1
1 1 ..5lb3 Game 41-
1 1 b3 Game 42-
1 1 .. .'ii'c7
1 1.. . ..td7 Game 40 -
19 ..tc1 Game 37
-
85
CHAPTER SIX I
3 cs 4 exd5 'ifxd5: Early
Deviations from the Main Line
. . .
86
3 . . . c S 4 e x d S 'ik x d S : Ea rly De viations from t h e Main L ine
1 0 . . . a6
10 ...'iixd4 and 10 ... .td7 are consid 2 7 .te3?
ered in Games 47 and 48 respectively. This just loses material. The last
1 1 iLf4 'ii'xd4 1 2 lll xd4 b5 1 3 iLe2 chance was 27 i.g3, although White's
il.b 7 1 4 lll b 3 uncoordinated pieces ensure that
The immediate 14 c4 led to speedy Black retains the advantage after
equality in Renner-Levitt, Augsburg 27 ...tt:Jc3 28 :b2 :as.
1989, after 14 ... bxc4 lS i.xc4 i.e7 16 27 . . . iLxe3 28 fxe3 lllc3 29 e4 iLxc4
.l::!.ac l tt:Jds 17 .tes 0-0 18 4Jb3 a5 19 0- 1
.l::!.fdl :fcs 20 .tbs a4 2 1 .l::!.xc8+ .l::!.xc8 After 30 i.xc4 lDxa2 3 1 i.xa2 Black
22 i.xa4 l:.a8 23 .tbs .l::!.xa2. With the can even round up the a6-pawn with
text move White plans to harass the 3 1 . . ..l::!.c 8, intending ... :c6. Some might
black bishop with ltJaS. say that the result was an apt punish-
87
The Fre n c h T a rra s c h
88
3 . . . c S 4 e x d S 'ii x d S : Ea rly De v ia t io n s from the Main L ine
ported by the two rooks, marches re .l::!.fdl l::txd7 18 .l::!.xd7+ tt:'ixd7, which he
lentlessly up the board. assesses as equal. If instead 16 ... tt:'ixd7
56 . . .:f 7 57 h6 l::th7 5a l:tg4 �c5 59 17 .l::!.fdl l:thd8 1 8 f4 White may be a
l::tg 7 :ha 60 h7 a2 61 �b2 1 -0 shade better.
89
The F re n c h Ta rra s c h
cxd4 5 exd5 'ii'x d5 6 iLc4 'ii'd 6 7 0-0 eventually lost. Instead White can win
lDf6 8 lDb3 lDc6 9 'ii'e 2!? with the brilliant 28 .l::!.xb7!! .l::!.xb7 29
tbg6+ @g8 30 1i'e8+ @h7 31 iixh8+
@xg6 32 .td3+ @gs 33 h4+ @xh4 34
'iic 8!, threatening the rook and also
mate with 3S g3+ @gs 36 f4+ @h6 37
'iih 8.
9 . . .iLe7
Also critical is 9 ...a6, after which
White has two options:
a) 10 a4 (m my opinion the inser
tion of a2-a4 and ...a7-a6 has helped
Black, who now doesn't have to 1 0 il.g5 0-0 1 1 l:.fe1 a6 1 2 :ad1 b5
worry about possibilities of .tbs or 1 3 il.d3 il.b7 14 c3 :tea 1 5 iLb1
tbbs) 10 ... .te7 1 1 g3 (1 1 .tgs es 12 'ii'c7 1 6 lDbxd4 lDxd4 1 7 lDxd4 :ad8
l:tfel .5lg4 13 h3 i.xf3 14 'iixf3 0-0 lS So finally White has recaptured the
i.h4 e4 was clearly better for Black in d4-pawn and a typical middlegame
Siklosi-Brinck Claussen, Copenhagen position has arisen: White's active
1988) 1 1. ..eS 12 t'.DgS 0-0 13 f4 and here pieces against Black's super-solid pawn
Black should play 13 ... i.g4! 14 tbxf7 structure.
i.xe2 lS tbxd6+ i.xc4 16 tbxc4 e4 1 8 'ii'd3 g6 1 9 'ii'h 3 lDh5!?
with an awesome looking centre.
b) 10 .l::!.d l! looks more testing, e.g.
10 ... bs 1 1 .td3 'iic7 12 a4 b4 13 i.c4
.tb7 14 tbbxd4 tbxd4 lS .l::!.xd4 i.cS 16
.l::!. h 4 hs 17 h3 tbds 1 8 .tgs .te7 19
ll.dl as 20 .tbs+ @f8 21 :c4 and
Black's king is awkwardly placed on
f8. Geller-Dolmatov, Moscow 1992,
continued 21...1i'd8 22 i.xe7+ 'iixe7
23 tbes @g8 24 tbc6 1i'f6 2S .l::!.xdS
exdS 26 tbe7+ @f8 27 .l::!.c7 .l::!.b 8.
see following diagram
20 il.h6? !
Here Geller played 28 t'.Dc8? g6!, and In this type of position White often
90
3 . . . c S 4 e x d S 'ik x d S : Ea rly De via tions fro m the Main Line
only gets 'one bite of the cherry'. A game drifts towards a draw.
chance of a sacrifice presents itself 30 . . . f3?!
with 20 lt:Jxe6!, when after 20 ... fxe6 Looks strong, but this pawn 1s
(20 ... .l::!.x dl 21 J::.xdl fxe6 22 'ifxe6+ picked up in the endgame.
@f8 23 l:td7 wins) 2 1 'ifxe6+ Black has 3 1 g3 a5 32 'ii'd 2 'ii'x d2 33 lll xd2
to be extremely careful, e.g. il.d5 34 a4 il.g5
a) 2 1...@f8? 22 .l::!.xd8 'ifxd8 23 i.h6+ After 34 ... i.c6 35 i.c2 i.c3 36 �e4
lt:Jg7 24 i.xg6!! hxg6 25 'ifxg6 ..tf6 26 ..5lb4 Black's can still play for the win
.l::!.xe8+ 'iixe8 27 'iixf6+ 'iif7 28 i.xg7+ with the bishop pair. Perhaps a draw
with a winning position. was all the Russian technician needed.
b) 2 1 ...@g7! 22 l::txd8 i.xd8 23 35 iLe4 iLxd2 36 iLxd5 �g7 37
i.h6+ @xh6 24 iixe8 with an unclear iLxf3 g5 38 � 1 f6 39 �e2 il.b4 40
imbalance of rook and two pawns iLe4 h6 4 1 �3 iLc3 42 �g4 il.b4 43
against bishop and knight. This may �f5 iLe 1 44 f4 gxf4 45 �xf4 il.b4
still be slightly better for White. 46 �5 iLc3 4 7 il.d5 il.b4 48 il.b3
20 . . . iLf6 21 'ii'g4 e5 22 lllf 5 :xd 1 iLc3 49 iLd 1 il.b4 50 iLh5 iLc3 51
23 l::t xd 1 .!:.dB 24 iLe3 lllf4 25 iLxf4 �e6 il.b4 52 �d5 Yz -Yz
exf4 26 llld4 b4
White's kingside demonstration has Game 50
got absolutely nowhere, and Black is Winsnes-Lein
just about to take over operations. In Gausdal 1990
particular, the pair of bishops will
make White suffer painfully in this 1 e4 e6 2 d4 d5 3 llld 2 c5 4 exd5
open position. 'ii'x d5 5 lllgf3 cxd4 6 iLc4 'ii'd 6 7 0-0
27 'ii'e 2 bxc3 28 bxc3 'ii'xc3 29 lllb3 lllf6 8 �b3 lllc6 9 :e 1
l::txd 1 + 30 'ii'xd 1
91
The Fre n c h Ta rra s c h
9 . . . il.d7 1 0 g 3 iLe7 1 1 iLf4 'ii'b4 1 2 her being slightly surprised and a little
'ii'd 3 0-0 1 3 iLc 7? ! annoyed when my opponent didn't
This i s probably too ambitious, al co-operate. Instead there came 8 ...eS 9
though it is quite surprising how 'ife2 ..5le7?! (after 9 ... ..5le6! it is not clear
quickly White's position deteriorates. how White can justify his pawn defi
I prefer 13 a3 'iib 6 14 lt:Jfxd4 lt:Jxd4 15 cit) 10 .l::!. dl 'iic7 1 1 ..tds ..tg4 12 'iic4
'iixd4 with a level position. @f8 13 l:Iel ..tf6 14 c3! and White
1 3 . . . iLdB 1 4 a3 'ii' e7 1 5 iLf4 il.b6 1 6 went on to win in Emms-I.Andersen,
lllfxd4 lllxd4 1 7 lll xd4 'ii'c 5! 1 8 J:tad 1 Cappelle la Grande 1992. 8 t'.De4 was
l:tac8 1 9 il.b3 'ii'h 5 20 'ii'e 2 lllg4 2 1 suggested to me by the American GM
f3 Ilya Gurevich, and it does seem a logi
cal way to exploit the omission of
.. .tt:Jf6.
92
3 . . . c s 4 e x d S 'ik x d S : Ea rly De viations from t h e Main Line
24 <it>f2 �8 25 h3 h5 26 �3 iLf5
27 g4 hxg4+ 28 hxg4 iLe4+ 29 <it>e3 a) 9 ... ..td7 10 0-0-0 0-0-0 1 1 lDbxd4
<it>e7 30 f5 a6 3 1 f6+ <it>d7 32 <it>f4 ltJxd4 12 lhd4 1i'b6 13 'fies 'fies 14
l::te8 33 e6+! .l::!.h dl with an edge to White in Palac
This sacrifice gives the white king a Touzane, Nice 1994.
decisive entry square, where it can b) 9 ... iib4+ 10 ..td2 1i'b6 1 1 0-0-0
support the f6-pawn. Black's battle ..td7 12 ..tgs 0-0-0 13 tijfxd4 ltJb4 14
against this monster of a passed pawn a3 ltJbdS lS i.xdS exdS 16 f3 i.d6 17
is doomed to failure. g3 and again White was slightly better
33 . . . fxe6 34 <it>e5 g5 35 f7 .l:.f8 36 in Rozentalis-Glek, Antwerp 1993 .
<it>f6 .l:.h8 37 tl:lxe6 .l:.h6+ 38 <it>xg5 1 0 0-0-0 b5 1 1 il.d3 iLe7 !
.tt. g6+ 39 <it>h5 .l:.xe6 40 f8if il.g6+ This is safer than 1 1...i.b7. Smagin
41 <it>g5 :xe 1 42 'ti'g7+ �e7 43 Levitt, Amantea 1993, continued 12
'ii' xg6 :ea 44 'ii' d3 b5 45 'ti'xd5+ 4Jbxd4 lDxd4 13 4Jxd4 0-0-0 ('castling
l::td6 46 'ti'xd6+ 1 -0 into it' has never been more apt)
Game 52
Anka-1.Almasi
Hungary 1997
1 e4 e6 2 d4 d5 3 tl:ld2 c5 4 exd5
'ii'x d5 5 tl:lgf3 cxd4 6 iLc4 'ii'd 6 7
'ii'e2 ! ? tl:lf6 8 tl:lb3 tl:lc6 9 il.g5
9 . . . a6
Black has no real route to comfort
able equality by giving the pawn back, 14 i.xbS!! axbS lS lDxbS 1i'b4 16
93
The Fre n c h Ta rra s c h
94
3 . . . c 5 4 e x d 5 'ik x d 5 : Ea rly D e viations from t h e Main Line
ttJb5 1i'b6 12 i.e3 i.c5 13 i.xc5 'iixc5 der to reach the diagram position be
14 1i'e3! 1i'xe3 15 fxe3 with an edge. low is 6 ...1i'd8 7 0-0 a6 8 tiJb3 'iic7,
1 1 il.b3 which avoids 8 tLle4.
1 1 i.xe6!? fxe6 12 tLixe6 i.xe6 13 7 0-0 a6 8 lll b3
1i'xe6+ is a dangerous piece sacrifice 8 tLle4!? may be a way for White to
which worked well for White in For exploit Black's move order, e.g.
ster-Vaganian, Biel 1994. After 8 ... 1i'c7 9 i.b3 i.e7 10 tLlxd4 tiJf6 1 1
13 .. . 1i'e7? 14 1i'c8+ 1i'd8 15 1i'xb7 %:!.e l with an edge for White.
Black's king didn't survive too long. 8 . . . 'ii'c 7
13 ... i.e7 would have been more stub
born.
1 1 . . .il.e7 1 2 :e 1 0-0 1 3 il.g5 il.d6
14 'ii'd 3 h6?
9 'ii'e2
The major alternative is 9 1i'xd4!?
Practical play suggests that White can
secure an advantage, e.g. 9 ... tiJc6 10
Black's development disadvantage 'iih4 i.d6 11 i.d2 ttJge7 (1 1...tiJf6 12
has reached serious proportions. i.h6!) 12 i.d3 i.d7 (12 ...ttJe5 may be
However, his position would still be stronger) and now:
playable but for this move, which al a) 13 lladl 0-0-0 14 tiJg5! tiJg6 15
lows a powerful sacrifice. 'iih3 tLlge5 16 i.e2 and White was bet
1 5 lll x e6! ! iLxe6 1 6 l::txe6 fxe6 1 7 ter, although he later lost, in Akopian
'ii'xg6 lll c6 1 8 :d 1 lll e5 1 9 'ii'xe6+ Dreev, Linares 1995.
<it>h8 20 :xd6 lllxf3+ 21 gxf3 :ae8 · b) 13 i.c3 f6 14 'iih5+ �f8 15 ttJc5
22 'ii'g 6 'ii'a 5 23 iLxh6 1 -0 ttJd5 16 ttJxd7+ 'iixd7 17 i.d2 and on
this occasion White converted his ad
Game 54 vantage in Nisipeanu-Priehoda, Buda
Emms-LB.Hansen pest 1996.
Bronshoj jubilee 1995 9 . . .il.d6 1 0 lll bxd4 llle 7
The main nuance in this line. This
1 e4 e6 2 d4 d5 3 llld2 c5 4 lllgf3 knight develops at e7 rather than f6.
cxd4 5 exd5 'ii'xd5 6 iLc4 'ii'd6 1 1 :d 1
In fact, the most common move or- 1 1 %:!.e l is a critical alternative, e.g.
95
The Fre n c h Ta rra s c h
96
3 . . . c S 4 e x d S Wk x d S : Ea rly De via tions from t h e Main Line
'>th2 'ii'e 5+ 46 <it>h 1 'We 1 + Yz - 1/:i ble is 9 ...tt:'ig4 10 0-0 tt:'ideS. Now 1 1
tt:'ixeS iixeS! is equal, while Svidler
Game 55 Savchenko, Kazan 199S, ended in a
Rublevsky-Beliavsky quick draw after 1 1 i.e4 tt:'ixf3+ 12
Novosibirsk 1995 tt:'ixf3 'fibs 13 h3 tt:'if6 14 i.b2 tt:'ixe4 lS
'ifxe4 f6 16 J::.adl lh-lh. Note that 13
1 e4 e6 2 d4 d5 3 llld 2 c5 4 exd5 i.b2 would have been met by
'ii'x d5 5 dxc5 iLxc5 6 lll gf3 lllf6 7 13 . ..tt:'ixh2!
il.d3 0-0 8 'ii'e 2 lll bd 7! 1 0 il.b2 b6 1 1 0-0-0 il.b7 1 2 :he1 ? !
Played by Anand against Kasparov. This turns out to be too pedestrian.
The knight is most useful here, as it Beliavsky recommends the direct 12
adds extra protection to the kingside g4, intending g4-gS, when 12 ... tt:'ixg4
and will not obstruct the black bishop fails simply to 13 i.e4, winning mate
when it fianchettoes on b7. rial. Now Black gets in first.
1 2 . . .a4 1 3 llle4 'ii'h 5 1 4 lllfg5 'ii'h 6
1 5 <it>b 1 axb3 1 6 axb3 iLa3 1 7
lllxf6+?
The start of a hallucination, based
on a horrendous oversight. 17 i.d4!
was stronger, after which Beliavsky
planned 17 ...tt:'idS!? 1 8 tt:'ixh7 .l::!.fc8 with
good compensation for the pawn.
1 7 . . . lllxf6
9 b3
Kasparov-Anand, Reggio Emilia
199 1 , continued 9 tt:'ie4 b6! 10 tt:'ixcS
'fixe s (10 ...tt:'ixcS is also okay for
Black) 1 1 i.e3 1i'c7 12 i.d4 i.b7 13
0-0-0 tt:'icS! 14 i.eS tt:'ixd3+ lS klxd3
1i'c4 and Black was fine . Indeed, the
game ended in a rare Kasparov defeat
with the white pieces.
9 . . . a5!? 1 8 iLxf6? 'ii'xf6 1 9 iLxh7+ <it>h8 20
Discouraging White from castling 'ii'h 5
queenside, although in this game Threatening mate in two, but ...
White will not be deterred. Also feasi- 20 . . . 'i¥b2 mate!
97
The Fre n c h Tarra s c h
Summary
Of White's alternatives in this chapter, perhaps the most dangerous are the early
'iie 2 ideas and the neglected 5 dxcS. It seems that further games are still required
before an accurate assessment can be made on the 6 ...'iid8 and 7... a6 line. '
5 lllgf3
5 dxcS - Game 55
5 . . . cxd4 6 iLc4 'ii'd6
6 ... 'iid8 7 0-0 (D)
7 ...0ie7 Game 53
-
7 0-0
7 'iie2 - Game 52
7 ... tllf6
7 ...0ic6 Game 51
-
7 ...a6 Game 54
-
9 %:!.el Game 50
-
1 1 iLf4 Game 46 -
7 0-0 8 lllc6
. . . 10 'ii'xd4
98
CHAPTER SEVEN I
3 cs 4 exd5 exd5:
. . .
99
The F re n c h Tarra s c h
the tiJb3-d4 lines, a tempo less fo r each Black can also play 14 ....l::!.d 8, which
side (see Game 60). may be stronger (see the next game).
7 . . . iLxc5
7 ...'iie 7+ crops up from time to
time, but it doesn't really cause White
any problems. For example, 8 iie2
'ifxe2+ 9 @xe2 i.xcS 10 tiJb3 i.b6 1 1
i.e3 i.g4 1 2 h 3 i.hS 1 3 i.xb6 axb6 14
tiJbd4 and Black's wrecked structure
gave White a definite plus in Tal
Korchnoi, Moscow 1973.
8 0-0 lll g e7 9 lll b 3 il.d6
100
3 . . . c s 4 e x d S e x d S : Main Line with 6 i. b S
101
The F re n c h Ta rra s c h
1 02
3 . . . c S 4 e x d S e x d S : Main L ine with 6 i. b S
103
The Fre n c h Ta rra s c h
31 . . . J:txf3
3 1.. .:f6 32 ltJxgS+ \tg6 33 .l::!.g8+
\th6 34 tt:Je8! is also hopeless.
32 gxf3 j_97 33 t2Jxb7 j_f6 34 j_c3
j_xc3 35 bxc3 j_xc4 36 t2Jd8 j_b5 A good move. Once the dark
37 t2Jxc6 j_xc6 38 l:te7 <it>g6 39 J:txa7 squared bishops are exchanged, the
j_xf3 40 c4 'itf6 41 r.Ia3 j_b 7 42 queen sits very nicely on this square.
l::t e3 h4 43 c5 j_d5 44 <it>h2 'itf5 45 Often Black follows up with ... a7-a5,
l::t e 8 j_b7 46 J:tg8 1 -0 planning ... a5-a4, which may loosen
White's control over the d4-square.
Game 59 1 5 a4 h5!?
Agnos-Lputian 15 ...a6 16 i.xc6 i.xf3 17 'ifxf3 bxc6
Linares 1996 18 'i'e3 probably keeps an edge for
White, who controls many dark
1 e4 e6 2 d4 d5 3 t2Jd2 c5 4 exd5 squares on the queenside.
1 04
3 . . . c S 4 e xdS e xd S : M a in L ine with 6 i. b S
105
The F re n c h Tarra s c h
White is a rook down, but the better for Black) 21...tbxf5 22 tbxf5
threat of b2-b4 followed by i.c4 is nfe8 23 i.d4 i.f8 24 .l::!.de l 'i!ld7 25 b3
extremely difficult to meet. The queen tbd6 and the players agreed a draw.
and knight once more prove to be an This is a good example of accurate
irresistible attacking force. If now play by Black. In the final position
29 ...'illcl 30 b4 tbe5 3 1 'i!lf6+ \tg8 32 White has absolutely nothing .
.l::!.xa5 the white rook also enters the
attack and Black cannot set up a de
fence. The move chosen by Lputian
also fails to stem the flow.
29 . . . J:tc7 30 b4 J:tf7 31 'ifh6+ <it>g8
32 'ifxh5 J:te6 33 bxa5 'iff6 34 j_c4
J:tfe7 35 t2Jxe7+ 1 -0
A powerful attacking display by
Agnos.
Game 60
Emms-Prandstetter
Barcelona 1993 1 3 J:te 1
13 i.e3 is probably more precise.
1 e4 e6 2 d4 d 5 3 t2Jd2 c5 4 exd5 Then after 13 ... a6 14 i.e2 tbxd4 15
exd5 5 l2Jgf3 t2Jc6 6 j_b5 j_d6 7 0-0 'i!lxd4 tbc6 the queen can retreat to d2.
tbge7 8 dxc5 j_xc5 9 ll'ib3 j_d6 1 0 That said, following 16 'i!ld2 .l::!. fe8 17
t2Jbd4 .l::!.adl .l::!.ad8 18 i.b6 i.c7 19 i.xc7
This position can also arise after 'i!lxc7 20 l:t.fe 1 h6 White had only an
7 ...cxd4 8 tbb3 tbge7 9 tbbxd4 (with infinitesimal edge in Karpov-Korch
one tempo less for either side). This noi, Moscow (game 16) 1974.
variation was given a thorough testing 1 3 . . . a6 14 j_e2 t2Jxd4! 1 5 'it'xd4 t2Jc6
in the 1974 Karpov-Korchnoi match, 1 6 'ifd 1 J:tfe8
the general verdict being that White's An exchange of minor pieces often
advantage was just too minute to cause favours the side battling against the
his opponent serious trouble. isolated pawn, as the weakness of the
1 0 . . . 0-0 1 1 c3 j_g4 1 2 'ifa4 'it'd7 IQP becomes more pronounced as the
12 ...i.h5 is a fully playable alterna endgame approaches. Here, however,
tive. Howell-Psakhis, Bled 1995, con Black is perfectly co-ordinated and
tinued 13 i.e2 a6 14 i.e3 'i!lc7 15 h3 even has a slight development advan
tba5! 16 .l::!.a dl l:t.ad8 17 l:tfe l h6 18 tage.
tbh4 i.xe2 19 l:txe2 tbc4 (once the 1 7 j_93 J:tad8 1 8 'ifc2 h6 1 9 J:tad 1
knight reaches this square, Black has 'it'c7 20 h3 j_h5 2 1 <it>f1 ? !
often solved his difficulties) 20 tbdf5 A ludicrous attempt to centralise
b5 2 1 'i!lc2 (2 1 'i!lxa6? l:ta8 22 'i!lxb5 the king for the endgame! Black could
l:tfb8 23 tbxd6 l:txb5 24 tbxb5 'i!lb7 is have exploited this immediately by
106
3 . . . c S 4 e x d S e x d S : Main Line with 6 i. b S
Game 61
Jansa-Votava
Prague 1993
1 e4 e6 2 d4 d5 3 t2Jd2 c5 4 l2Jgf3
t2Jc6 5 exd5 exd5 6 .ib5 .id6 7 dxc5
j_xc5 8 0-0 tbge 7 9 t2Jb3 j_b6
24 . . .'ir'b6?! A common alternative to the usual
This leads to problems. 24 ... d4! 25 9 ...i.d6.
cxd4 ltJxd4 26 l:t.xe8+ l:t.xe8 27 ltJxd4 10 J:te 1 !
i.xd4 is the simplest way to draw. The plan involving l:tel and i.e3 is
25 l:txe8+ l:txe8 26 l:td2 d4 27 cxd4! by far the most popular idea for
Black was hoping for 27 ltJxd4 White. If truth be told, most Black
ltJxd4 28 cxd4 'ifb4. Around this stage players shy away from 9 ... i.b6 .for
I realised that I actually had some real precisely this reason. Other ideas pale
winning chances. by comparison. For example, 10 ltJbd4
27 . . . 'ir'c7 28 J:tc2 'ir'f4 29 'ir'g4! 'ir'xg4 0-0 1 1 c3 i.g4 12 i.e2 ctJxd4 13 ltJxd4
30 hxg4 t2Jb4 i.xe2 14 ctJxe2 ctJc6 15 ltJf4 d4 16 cxd4
The best move. 30 ... ltJxd4 3 1 .l::!.c7! i.xd4 gave Black no problems at all in
.i::t b 8 32 i.c4 is clearly better for Renet-Korchnoi, Paris (rapidplay)
White, while after 30 ... i.xd4 White 1990.
doesn't play 3 1 .l::!.xc6 bxc6 32 ltJxd4 10 ...0-0 11 j_e3 j_g4 1 2 j_xb6 axb6
.l::!.d8, but instead 3 1 i.xa6 ltJb4 32 .l::!.c4 The sharper 12 ...'ifxb6 will be ex
bxa6 33 .l::!.xb4. amined in the next game. There is
31 J:td2?! nothing really to this position apart
As both players were entering into from Black's dodgy pawn structure,
big time-trouble, the inevitable errors which gives White a small but obvious
creep in. 3 1 .l::!. c3 is stronger here, as advantage.
31...ctJxa2 can then be met by 32 .l::!.c7. 1 3 j_e2 'ii'd6 1 4 c3 l:tfd8 1 5 t2Jfd4
Now 3 1 ...ctJxa2 should have been j_xe2 1 6 J:txe2 l2Jg6 1 7 t2Jf5 'ir'f6 1 8
107
The Fre nch Ta rra s c h
1 08
3 . . . c s 4 e xdS e xd S : Main L ine with 6 i. b S
Tondivar was actually repeating one no choice as the knight has no useful
of his previous games, which had gone retreat square.
20 l:t.xb7 'ifxb2 21 l:t.xa7 h5 22 \t>g2 27 . . .'ir'xc3 28 J:ta8 1 -0
ltJcl 23 ltJd4 'ifc3 24 'iff4 'ife l 25 'ife3 Since 28 ...'ifc7 runs into 29 'if e8+. A
.l::!. e 8 26 'ifxel .l::!.xe l 27 %:!.al 1-0 Jansa powerful performance from Dvoirys.
Tondivar, Ostend 1993. No doubt he The onus is on Black to come up with
had some improvement lined up, al a substantial improvement here.
though I can't see anything special. In
any case, Dvoirys got his new move in Game 63
first. T. Ravi-Dolmatov
20 . . .'ir'g5+ 2 1 <it>f1 'ir'c 1 + 22 <it>g2 Calcutta 1996
'ir'xb2
Now that the king is more strongly 1 e4 e6 2 d4 d5 3 llld2 c5 4 lllgf3
placed on g2, White would have an lllc6 5 exd5 exd5 6 j_b5 cxd4
swered 22 ...'ifg5+ with 23 'ifg4. This is a clever attempt to steer the
23 �xb7 'ir'c3 24 l::tx a7 lll b4 25 J:ta4! game into lDbd4 lines, which would
White has kept the extra pawn, and arise after 7 0-0 .td6 8 lDb3 lDge7 9
now initiates a surprising attack which lDbxd4. At the moment it is not im
is immediately decisive. mediately obvious how White should
25 . . . 'ir'c5 26 lllf5 best avoid this transposition.
7 'ii'e2+
After 7 lDxd4 .td7 8 'ife2+ 'ife7 9
lD2b3 'ifxe2+ 10 lDxe2 tDf6 1 1 i.e3 a6
12 i.xc6 i.xc6 13 0-0-0 0-0-0 14 .td4
ltJd7 15 ltJg3 .l::!.g8 16 f3 g6 17 fDe2 .td6
the position was equal in Brodsky
Dolmatov, Novgorod 1995. The d
pawn is difficult to attack in the ab
sence of a light-squared bishop.
7 . . .'ir'e7
26 . . .J:tbS
Other moves also lose:
a) 26 ...ltJc6 27 nc4.
b) 26...ftJds 27 'ifxd5 'ifxds 28 lDe7+.
c) 26 . . .fDxc2 allows the prettiest fin-
ish, as 27 lDe7 + \ti h8 28 'ifxh7 +! \t>xh7
29 l:t.h4+ leads to mate.
27 c3!
27 .l::!. a 8 allows the defence 27 ... 'iff8,
so White deflects the queen. Black has
109
The Fre n c h Ta rra s c h
1 10
3 . . . c S 4 e x d S e x d S : Main L ine with 6 i. b S
Summary
In the main line with 9 ... .td6 it seems that White has reasonable chances for an
advantage with either 1 1 .td3 or the older 1 1 .tgS. 9 ....tb6 still has a poor repu
tation and this line needs some major surgery before it becomes playable. If
Black really has to capture with 12 ... axb6 (as in Game 61) then I suspect we
won't see much of this line in future. The variation with 6 ... cxd4 (Game 63) cer
tainly merits further investigation, but even here Black players may have to ac
cept that White can always make an easy draw.
7 dxc5
7 0-0 cxd4 8 tt:Jb3 0-0 9 tt:Jbxd4 - Game 60 (by transposition)
7 . . .j_xc5 8 0-0 l2Jge7 9 t2Jb3 j_d6
9 ... .tb6 10 �e l 0-0 1 1 .5le3 .5lg4 12 .5lxb6 (D)
12 ...axb6 Game 61
-
12 ...'ifxb6 Game 62
-
10 �e1
10 tt:Jbd4 - Game 60
1 0 ... 0-0 1 1 j_d3 (D)
1 1 .tgS - Game 59
11 ... h6 1 2 h3 t2Jf5 1 3 c3 'it'f6
13 ... .tc7 Game 58
-
15 'it'd3 Game 56
-
111
CHAPTER EIGHT I
3 cs 4 exdS exdS:
. . .
1 12
3 . . . c s 4 e x d S e xd S : Fifth Mo ve A lt e rn a ti v e s
0-0 ile7 8 dxc5 ilxc5 9 tt:Jb3 il..b6 10 White after this game, although
il..g5 0-0 11 c3 l:t.e8 12 il..h4 h 6 13 l:t.el Black's play was far from inspiring.
g5! 14 il..g3 tt:Je4 15 tt:Jfd4 f5! 16 Ji..h5 1 2 j_93 t2Jfe4
l:t.f8 and Black had a very active posi Apart from the challenging 12 ... a4,
tion in Onischuk-Kramnik, Tilburg which is the subject of the next game,
1997. Black has also tried:
6 . . .j_d7 7 j_xd7+ t2Jbxd7 8 0-0 j_97 a) 12 ... b5 13 il..d4 tt:Je6 14 il..e5 tt:Jg4
9 dxc5 t2Jxc5 1 0 J:te 1 15 tt:Je3 lDxe3 16 .l::!.xe3 'ifd7 17 'i'd2
.i::!.ad8 18 l'!dl tt:Jc5 19 il..c3 Ji..f8 20
.l:txe8 'ifxe8 21 il.. d 4 tt:Je4 22 'ifaS and
Black's pawns were looking rather
vulnerable in Yurtaev-Ulibin, Russia
1997.
b) 12 ... a6 (too slow) 13 c3 l:tc8 14
'ifc2! 'ifc7 15 .l::!.a dl b5 16 il..d4 and
once more White has reached an ideal
set-up, as in Kasparov-Short, London
(rapidplay) 1993.
1 3 c3 'ifd6 14 l2Jg3 j_f8 1 5 'ifc2
'ifa6 1 6 J:ted 1 J:tad8 1 7 l2Jg5! t2Jxg5
The start of a logical plan which After 17 ...'ifg6 18 lD5xe4 dxe4 19
was introduced by the Swiss GM Lu .l::!.xd8 l:t.xd8 20 ilxc5 ilxc5 White
cas Brunner. White's idea is quite doesn't play 21 lDxe4? because of
clear: He plans tt:Jfl and il..e3-d4, fol 21...l:t.e8 22 l:t.el f5 winning a piece,
lowed by tt:Je3 and possibly 'ifd3 and but 21 'ifxe4!, which leaves him a clear
l:t.adl. The white pieces would then be pawn up.
very harmoniously placed and the d5- 1 8 j_xg5 J:td7 1 9 j_93 b6 20 'iff5
pawn would also come under pres g6?!
sure. This concedes a slight weakness on
10 tt:Jd4 and 10 tt:Jb3 are considered f 6, which White is able to utilise.
in Games 66 and 67 respectively. 20 ...'ifb7 would have kept White's
1 0 . . . 0-0 1 1 t2Jf1 l::te 8! plus to a minimum.
The stem game Brunner-Atalik, 21 'it'g4 t2Je6 22 j_d4! 'it'b5 23 l2Jh5
Kecskemet 199 1, varied with j_97 24 j_f6 d4?
1 1 ...tt:Jce4?! 12 il..e 3 .l::!.c 8 13 il..d4 il..d6 24 ... il..f8 would have been a better
14 lDe3 il..b8 15 'ifd3! l1e8 16 .l::!. adl defence, although after 25 ile5 ile7 26
'ifd6, and now instead of 17 g3?! .l::!. d2 the weaknesses around the black
White should have just played 17 king still give White a clear advantage.
il..xf6! tt:Jxf6 18 c4 with a big advan After 24...d4? White reaches a very
tage, as the d-pawn drops. It was pleasant endgame.
hardly surprising that the idea begin 25 j_xe7 'ifxh5 26 'ifxh5 gxh5 27
ning with 10 .l::!.e 1 became popular for j_f6 d3 28 J:td2 t2Jf4 29 <it>f1 l:.e2 30
1 13
Th e Fre n c h Ta rra s c h
J:tad 1
1 14
3 . . . c S 4 e x d S e x d S : Fift h Mo ve A lternatives
.l::!.e2 tl'ie4 16 tl'ie3 tl'i6g5 White has to J:td4 'ifc6 29 J:ted 1 J:te8 30 J:t 1 d3
play with care just to equalise, e.g. 17 l2Jg5 31 'it'd1 l2Jge4 32 J:txc4 dxc4 33
tl'ixgS (17 'ifxds?! i.xe3 1 8 'ifxd8 l::td4 b5 34 j_d2 t2Jc5 35 j_e 1 t2Jfe4
i.xf2+ 19 l:t.xf2 .l::!.axd8 20 l:t.e2 tl'ixf3+ 36 'iff3 'it'e6 37 l2Jg3 g6 38 t2Jxe4
2 1 gxf3 l:t.xeS 22 fxe4 fS wins a pawn) t2Jxe4 39 j_h4 h6 40 h3 <it>g7 41 <it>h 1
17 ... 'ifxgS 1 8 i.d4 tl'ixf2 19 .l::!.xf2 i.xd4 'it'c6 42 j_d8 l::te6 43 <it>g 1 'it'e8 44
20 'ifxd4 'ifxe3 2 1 'ifxdS .l::!.e7 22 l:t.dl .l::.d 5 l2Jg5 45 'iff4
h6 and the players shook hands on a
draw in Rozentalis-M.Gurevich, Bel
fort 1997.
1 3 . . .'it'd7 14 l2Jd4 a4 1 5 a3
White feels obliged to prevent the
possible ... a4-a3, but now Black ob
tains a bind on the light squares in the
centre and on the queenside.
1 5 . . .t2Jce4 1 6 'it'b5 'ifc 7 ! 1 7 l:tad 1
tbd6 1 8 'it'd3 t2Jc4 1 9 j_c 1 j_c5 20
t2Jf5 t2Jg4 2 1 l2J5e3 j_xe3 22 fxe3
t2Jf6
Black has lost control and no longer
possesses any real advantage. Now
45 ... .l::!.xe3 loses to 46 i.f6+ \t>h7 47
.l::!.d 8, so Black bails out.
45 . . .t2Je4 46 j_c7 Yz - Yz
Game 66
Karpov-Bareev
Linares 1994
1 e4 e6 2 d4 d5 3 t2Jd2 c5 4 exd5
exd5 5 t2Jgf3 t2Jf6 6 j_b5+ j_d7 7
Without making any obvious blun j_xd7+ t2Jbxd7 8 0-0 j_97 9 dxc5
ders, White has managed to reach t2Jxc5 10 t2Jd4 'it'd7!
quite a hideous position. Both c4 and The queen often finds itself parked
e4 are excellent outposts for the black on a light square in this line. Here it
pieces, while White has been saddled patrols the sensitive fS-square, which
with weak pawns on e3 and b2. Given the knight on d4 was eyeing.
the way things have gone so far, it is 1 1 t2J2f3 0-0 1 2 j_f4
quite a miracle that Gurevich survived The main alternative is 12 tl'ieS 'ifc8!
this position. and now:
23 'it'd4 t2Jd6 24 'iff4 J:te4 25 'it'g3 a) 13 'iff3 .l::!. e8 14 tl'ifS tl'ice4 15 c3
J:tc4 26 c3 'it'b6 27 'iff3 t2Jde4 28 and here not 15 ... i.d8?! 16 tl'ig4 .l::!.e6
1 15
The Fre n c h Tarra s c h
1 3 . . .j_fB
An early game in this line went
13 ... CL'ice4 14 CL'ie5 'ifd8 15 CL'id3! lk8 16
c3 i.f8 17 'ifb3 and the knights were
very nicely placed on d3 and d4, en
suring White an edge in Adams-Short,
English Championship 199 1 .
1 4 t2Je5 'ir'a4!?
Naturally Black can also consider 35 . . .j_a7?? 36 J:txd8 mate
14...'ifc8 and 14 ...'ifd8. A shocking finish and a reminder
1 5 c3 'ir'a6 1 6 'ir'e2 'ir'xe2 1 7 J:txe2 that it happens to the best of us! After
j_d6 such a disaster it only seems fair to
Allowing a neat trick, although it is include an instructive Bareev victory.
not clear that this changes the assess
ment that much. White has a minus Game 67
cule plus, which shouldn't really be l.Gurevich-Bareev
enough for victory. Biel 1993
1 8 t2Jd7! j_xf4
1 8.. Jhe2? 19 CL'ixf6+ gxf6 20 i.xd6 1 e4 e6 2 d4 d5 3 t2Jd2 c5 4 exd5
1 16
3 . . . c5 4 exd5 e x d 5 : Fifth Move A l ternatives
1 1 .i.f4
Another plan is 1 1 tt:Jfd4 'it'd7 12
'it'f3, planning to plonk a piece on the
f5-square. White then has two options 33 . . ..i.g3 ! !
after 12 . . 0-0:
. A brilliant move which exploits the
a) 13 'it'f5 g6! (Black doesn't fear a temporary disharmony in the white
queen exchange) 14 'it'xd7 tt:Jxd7 and ranks.
the d-pawn is quite safe. 34 c4
b) 13 tt:Jf5 i.d8! 14 i.e3 g6 15 tt:Jg3 Other moves are also bad for
ne8! 16 c3 a5 17 tt:Jd4 .l::!.a6 18 l:.adl White:
tt:Jg4 19 tt:Jc2 .l::!.f6 and Black had a a) 34 tt:Jxg3 .l::!.cd8! 35 'it'f5 'it'c4+.
strong initiative in Kotronias-Psakhis, b) 34 fxg3 .l::!.cd8 35 'it'f5 'it'c4!
Chalkidiki 1992. c) 34 .l::!.b2 is probably the best of a
1 1 . . .0-0 1 2 'it'd3 lies 13 .i.e5!? 'it'd7 bad bunch, although 34 ... i.xf2 is still
14 a4 a6 1 5 l:lfd 1 l:tac8 1 6 c3 .i.d8! grim for White.
Once again we see this clever move. 34 . . . l:tcd8 35 'iff5 'it'xc4! 36 l:le3
From d8 the bishop may shoot to c7 36 .lhc4 .l::!.d l+ 37 lDe l tt:Jd2+ 38
or b6, where it applies pressure on the �gl l:txel is mate.
central squares. There now follows a 36 . . ..i.bS 37 g3 'it'xb4 38 <it>g2 t2Jd6
117
The Fre n c h Ta rra s c h
3 9 'it'd3 J:txe3 4 0 'it'xe3 l:te8 41 'it'd3 sequence Black can also play the
'it'xa5 slightly strange looking immediate
5 ... c4. The drawback to this idea is
that after 6 b3 he is forced to exchange
pawns, as 6 ... bS walks into 7 a4. After
6 ... cxb3 7 axb3 ..5lb4 8 t'DeS t'Df6 9 ..5ld3
0-0 10 0-0 ..5lc3 1 1 l:t.a4 ..5ld7 12 t'Dxd7
t'Dbxd7 13 t'Db l! ..taS 14 ..td2 .tel 15
t'Dc3 White has unravelled successfully
and the bishop pair ensures an edge, as
in Svidler-Korchnoi, Groningen 1996.
1 18
3 . . . cS 4 e x d S e x d S : Fifth Mo ve A l tern a tives
1 19
The Fre n c h Ta rra s c h
8 j_d3
8 il..gS is the subject of the next
game.
8 . . . 'it'e7+!
This annoying queen check is the
reason why 8 il..d3 is not seen very
much. White has to choose between
losing a move with 9 ile2, or an un
welcome exchange of queens.
9 'ife2 tt:Jc6 1 0 c3
An early indication that Black has
no problems here was the game Mata
2 0 . . . j_f6 2 1 'it'g4+ � c 7 22 'iff4+ novic-Korchnoi, Wijk aan Zee 1968.
<it>c8 23 j_xf6 t2Jxf6 24 'ifxf6 'it'xc2 After 10 0-0 il..g4 11 h3 il..h S 12 il..f4
25 'iff5+! 1 -0 'ifxe2 13 il..xe2 tt:Jf6 14 c3 0-0 15 fHel
25 ...'itt c l 26 l:t.e7+ or 25 ...l:t.d7 26 l:t.fe8 16 g4 il..g6 17 il..fl .l::!.e4 18 .l::!.xe4
.l::!.e 8+ both win the queen. A powerful ilxe4 19 ilg2 hs 20 gs tt:Jd7 2 1 .l::!. dl
performance by Geller, but we await tt:Jf8 22 il..e 3 il..xe3 23 fxe3 tt:Je6 24 h4
more tests with 9 ... c3 ! l:t.e8 Black had taken over the initiative
and went on to win.
Game 69 1 O . . ,j_g4 1 1 0-0 'it'xe2 1 2 j_xe2 t2Jf6
Donchev-Eingom This endgame holds absolutely no
Debrecen 1992 worries for Black, whose active pieces
more than offset the weakness of the
1 e4 e6 2 d4 d 5 3 t2Jd2 a6 4 l2Jgf3 dS-pawn.
c5 5 exd5 exd5 6 dxc5 j_xc5 7 t2Jb3 1 3 h3 j_h5 1 4 g4 j_g6 1 5 t2Jfd4 h5
j_a7 1 6 <it>g2 hxg4 1 7 hxg4 t2Je5 18 f3
Here Black utilises the al-square 0-0-0 1 9 J:tf2 l:tde8 20 j_f4 t2Jc4 2 1
that was vacated by ... a7-a6. The re j_d 1 t2Jd7 2 2 j_c2 j_xc2 2 3 J:txc2
treat to this haven is obviously the
most natural move, but there is also
nothing wrong with 7 ... il..e 7, a favour
ite of Dolmatov. An example of
Black's chances is shown in the game
Einarsson-Dolmatov, Reykjavik 1988,
which went 8 il..d3 tt:Jf6 9 0-0 0-0 10
liite l tt:Jc6 11 c3 il..g4 12 il.. gS (12 h3
Jl.. h s 13 il..gS 'ifd6 14 .l::!.e3 looks more
testing) u ... h6 13 il.. h4 'ifb6 14 ilc2
tt:Je4! 15 'ifxdS?! ilxf3 16 gxf3 il..xh4 17
'ifxe4 ilxf2+ 1 8 'itifl g6 and White's
king didn't last many more moves. 23 . . .j_bS!
120
3 . . . c5 4 e x d5 e x d 5 : Fifth Mo ve A lternatives
Game 70 Game 71
Van der Wiel-Seirawan Rublevsky-Bareev
Biel 1985 Rubinstein Memorial 1997
121
The Fre n c h Ta rra s c h
A very direct move. Black simply lbc4 20 i.d4? i.a3!! 21 bxa3 'ii'd6 22
wishes to eliminate the knight on b3 lbc5 lbe4 with favourable complica
and regain the c5-pawn. If Black pre tions for Black, as in Yudasin-Psakhis,
fers to play in sacrificial mode then Beersheva 1993.
1 1 . ..lbg4 is worth considering. In the 1 5 a4 ltJxc5 1 6 ..ib5 lDxb3+ 1 7 cxb3
game Van Wely-Ehlvest, Biel 1993, 'ir'c7+ 1 8 <it>b1
White quickly gained a substantial White now has doubled pawns on
advantage after 12 0-0 a5 13 a4 i.f6? 14 the queenside, and the d5-pawn is also
c3 lba6 15 i.b5 lbc7 16 lbbd4 lbxb5 passed, as well as isolated. On the
17 axb5 'ii' c7 18 b4 i.xd4 19 b6 'ii'b 8 other hand, the white king is well pro
20 cxd4 axb4 21 'ii'd2. However, tected now, and he can hope to use his
13 ... lba6! is a stronger move. After 14 two bishops in an open position.
i.xa6 l::!.xa6 Black's two bishops give 1 8 . . Jled8 1 9 ..id4 ltJe4 20 f3 lDc5
him definite compensation for the 21 ..ie5 ..id6
pawn. 2 1...'ii'b 6 allows 22 i.xg7, so Black
1 2 lDfd4 lDbd 7 1 3 0-0-0 has to enter an inferior endgame,
The main move here, although the where the bishop is much stronger
quieter 13 0-0 must not be underesti than the knight. Rublevsky, however,
mated. Here Black should probably makes a few slight inaccuracies and
continue with 13 ...lbxc5, as 13 ...i.xb3 this allows Bareev to hold the draw.
was met by the zwischenzug 14 c6! in 22 l:.xd5 ..ixe5 23 'ir'xe5 'ir'xe5 24
the game Kosashvili-Sjodahl, Arnhem llxe5 lDxb3 25 <it>a2
1989, when after 14 ...bxc6 15 axb3 25 l::!.e7 looks like a better winning
i.c5 16 'ii'd l 'ii'b 6 17 l::!. a4 a5 18 h3 lbf8 try.
19 'ii'f3 Black's weak queenside pawns 25 . . .ltJd4 26 llc 1 llab8 27 l:.c7 ltJe6
gave White an edge. 28 l:.d7 l:.dc8 29 llf5 f6 30 l:.d6 ltJc5
1 3 . . . ..ixb3 1 4 lDxb3 31 g4 b6 32 h4 <it>f7 33 g5 lld8 34
Naturally 14 axb3 is also possible, llc6 l:.dc8 35 gxf6 l:.xc6 36 ..ixc6
although this may give Black more gxf6 37 llh5 l:.h8 � - �
chances as the b3-pawn can be a target
for the lunge ... a6-a5-a4. Van Wely Game 72
Brenninkmeijer, Holland 1993, con Moya-Roldan
tinued 14 ... i.xc5 15 'ii'f3 lbe5 16 'ii'h3 Correspondence 1984
'ii'b 6 17 lbf5 i.xe3+ 18 lbxe3 a5 19
lbxd5 lbxd3+ 20 'ii'xd3 lbxdS 21 'ii'xd5 1 e4 e6 2 d4 d5 3 lDd2 c5 4 exd5
a4, when White is a pawn up but faces exd5 5 ..ib5+ ..id7 6 'ir'e2+ 'ir'e7?!
some tricky moments on the queen The endgame reached after this
side. move is known to be comfortably
1 4 . . . a5 better for White.
Or 14 ...lbxc5 15 lbd4 (15 'ii'f3 is 7 ..ixd7+ ltJxd7 8 dxc5 ltJxc5 9 lllb3
probably stronger) 15 ...lba4!? 16 c3 'ir'xe2+ 1 0 ltJxe2 lDxb3 1 1 axb3 ..ic5
.Uc8 17 i.c2 lbb6 1 8 'ii'd3 g6 19 lbb3?! 1 2 ..id2
122
3 . . . c 5 4 e x d 5 e x d 5 : Fifth Mo ve A l tern a t i ves
1 23
The Fre n c h Tarra s c h
Summary
Despite a variety of attempts against 5 tbgf3 lbf6, Black's move has stood firm
and the onus is clearly on White to produce something new. The variation 5 ... a6
6 .ie2 c4!? is especially good for Black players determined to go for the full
point, as many of White's most promising lines involve sacrificing material. 5
.ibS+ continues to be less popular than 5 lbgf3, as it seems that Black has suffi
cient resources after either 5 ... .id7 or 5 . . lbc6 6 'ii'e2+ .ie7.
.
6 ...'ii'e7 Game 72
-
5 ...lbc6
6 lbgf3 - Chapter 7
6 'ii'e2 Game 73
-
5 . . .lDf6
5 ... a6
6 .ie2 Game 68
-
8 .igS Game 70
-
6 i.. b 5+ i.. d 7 7 i..x d7+ lDbxd7 8 0-0 i..e7 9 dxc5 ltJxc5 (D) 1 0 lle1
10 lbd4 - Game 66
10 lbb3 - Game 67
10 ... 0-0 1 1 lDf1 l::te8 1 2 i..e3 ltJfe4
12 ... aS Game 65
-
1 3 c3 -Game 64
124
CHAPTER NINE I
3 . . . cs 4 ltJgf3
1 25
The Fre n c h Ta rra sc h
126
3 . . . c5 4 lb g f3
------�
-------
ttJ';'!ii
�:--:=� '
(9 ...lDf6 !?) 10 i.b4+ 1 1 <tie2 'ii'a5
12 i.e3 i.e7 1 3 'iixd2+ 14 i.�d2,
as m . Van der Wi·el-Ehlvest, Hanmge
1990.
8 i..d3 'iic7
9 � "' 2 .>ild6
Black can als develop the kmght
.
first, whi�h has th advantage of tak
ing the sung :;_ut o; White's e4-e5. For
examp le, 9 . . .'1.Je7
lilf lO 0-0 t2Jg6 1 1 ltJf3
i.d6 12 :el 4 13 .hf4 �� xf4 14 c4
;nd h7 is droppmg.
1 6 c4 f5?
7 lllxc6 bxc6
7. . i.xc6 can lea_d to an endgame
. Blac k' s spht pawns on the
Where
. .
queenside · v white a minute advan .
tage, e.g. � �xc6+ bxc6 9 c4 dxe4 Thmgs were already becoming
127
The Fre n c h Ta rra s c h
Game 76
Popovic-P .Nikolic
Yugoslavia 1991
1 e4 e6 2 d4 d5 3 ltJd2 c5 4 lDgf3
lDf6 5 exd5
Here 5 e5 lbfd7 6 c3 lbc6 7 i.d3
transposes to lines considered in
Chapter 3, Games 27 and 28. After 5
exd5 Black can obviously play
5 ...exd5, converting to Games 64-67, 14 1Ld5!
but here Black chooses another en 14 i.d3 i.b7 would not have pre
tirely playable move. sented Black with any problems, but
5 . . . ltJxd5 6 lDb3 now White can steer the position into
an endgame, where his control of the
d-file will be the most important fac
tor. Note that 14 ... :b8 15 i.c6 l::!.b6
fails to 16 lbe5!
14 . . . ltJxd5 1 5 'ir'xd5 lla7 1 6 'ir'd6
'ir'xd6 1 7 llxd6 llc7 18 ltJe5 lDb8
This looks rather ugly, but
18 ...lbxe5 19 :xe5, with the idea of
llc5, also favours White.
1 9 ltJa5 f6 20 ltJd3 e5 21 lDb4 <it>f7
22 l:.ed 1 1Le6 23 c3 1Lc8 24 g3 h5
25 h4 lle8 26 lDd5 lla7 27 lDb4 lic 7
128
28 : 1 d2 :ea 29 :da :ea 30 :ada lDbca 8 c3 0-0
:ea 3 1 :da :ea 32 :xe8 <it>xe8 33
:da :a7 34 :ba lDd7 35 l:.ea+ �7
3a :ca :as 37 lDd5 e4?
Black has been forced to defend a
grim endgame for quite a while, and
finally cracks under the pressure.
37 ...:b8 was the only way to hang on.
38 :c7
9 exd5
The other way to play this position
is to maintain the pawn on e4. Rozen
talis-P.Nikolic, Moscow 1994, contin
ued 9 'ii'e2 lbg6 10 lbb3 dxe4 (perhaps
10 ....i.a7!?, maintaining the tension, is
more appropriate) 1 1 i.xe4 i.d6 12
i.e3 'ii'c7 (12 ...lbf4? loses 13 i.xf4
38 . . . <it>ea i.xf4 14 i.xh7 + 'it>xh7 15 'ii'e4+) 13
This loses a piece, but 38 ... 'it>e8 fails :adl :d8 14 'ii'c4! i.d7 15 i.xg6 hxg6
to 39 :xd7, while after 3 8 ...'it>f8 39 16 lbg5! and White had drummed up a
lbc6 Black is completely tied up. menacing initiative.
39 :xd7 1 -0 9 . . . exd5
White wins the rook on a8 after 9 ...lbxdS is also very reasonable, af
39 ... 'it>xd7 40 lb b6+ or 39 ... .i.xd7 40 ter which White should probably con
lbc7+. tinue with 10 lbe4. After 9 ...exd5 we
have actually reached a position simi
Game 77 lar to that discussed in Chapter 7 (see
Wolff·Yermolinsky Games 56-58). The difference here is
USA Championship 1993 that Black has committed himself to
...a7-a6, which is only a semi-useful
1 e4 ea 2 d4 d5 3 ltJd2 aa 4 lDgf3 move.
c5 1 0 lDb3 1'.da 1 1 :e 1 ha 1 2 h3 1Lf5
This position can also be reached After 12 ...lbf5, which is normal
via the move order 3 ... c5 4 lbgf3 a6. If procedure in Games 56-58, Black's
now 5 exd5 exdS we have reached the ...a7-a6 proves to be a serious loss of
standard position of the ... a7-a6 line time. White can simply play 13 i.c2
(see Chapter 8, Games 68-70) . i.e6 14 'ii'd3, when the threat of g2-g4
5 dxc5 1Lxc5 a 1Ld3 ltJe 7 7 0-0 is hard to deal with.
129
The Fre n c h Ta rra s c h
Game 78
Kuijf-Uhlmann
Amsterdam 1990
1 4 lDbd4
14 i.cS!? 'ii'd7 15 i.xfS lbxfS 16 1 e4 e6 2 d4 d5 3 ltJd2 c5 4 lDgf3
'ii'd3 is a sensible suggestion for White, ltJc6 5 ..ib5
when the pressure on d5 ensures a An interesting bid to avoid the
small edge. normal lines that arise after 5 exdS
1 4 . . . ..ie4 1 5 'ir'c2 ltJg6 1 6 ..ixe4 exdS (see Chapter 7).
dxe4 1 7 ltJd2 ltJxd4 1 8 ..ixd4 f5! 5 . . .dxe4
The nature of the position has sud This is the theoretical choice, al
denly changed. Black now has a po though 5 ... a6 also doesn't look bad,
tentially dangerous pawn roller on the e.g. 6 i.xc6+ bxc6 7 0-0 lbf6 8 es lbd7
kingside. Wolff now decided to grab a 9 c4 i.e7 10 dxcS lbxcS 1 1 lbd4 'ii'b6
hot pawn on b7, but Alex 'the Yermi 12 'ii'g4 0-0 13 lb2f3 fS with an unclear
nator' Yermolinsky utilised his initia position in Sutovsky-Psakhis, Haifa
tive to the full. 1996.
1 9 'ir'b3+ �h7 20 'ir'xb7 Note that 5 ... cxd4 6 lbxd4 simply
transposes to Game 75.
6 lDxe4 ..id7
130
3 . . . c s 4 li:J g f3
20 . . . ltJe7!
This regrouping manoeuvre spells
the beginning of the end for White.
Black wins a pawn after both 2 1 .Ufdl
l':.xd6 22 cxd6 lhd6 23 .Uxd6 'ii'xd6
7 . . .'ir'a5+ 8 ltJc3 cxd4 9 ltJxd4 ..ib4 and 2 1 i.d4 lbfS 22 l:.xd7 lhd7 23 c3
1 0 0-0 ..ixc3 1 1 bxc3 ltJge7 1 2 eS, while White's chosen move also
ltJxc6 ..ixc6 1 3 ..ixc6+ !Lixc6 1 4 c4 sheds a pawn.
0-0 1 5 °ir'b1 'ir'c7 1 6 c5 llfd8 2 1 ..ig5 llxd6 22 cxd6 llxd6 23 ..if4
Black has no weakness, whereas 'ir'b6! 24 'ir'e5 lld7 25 ..ie3 'ir'c7 26
White's doubled c-pawns are always a °ir'b2 lDf5 27 ..ib6 'ir'c6 28 g3 h5 29
cause for concern. Optically things do h4 lld2 30 'ir'b4 ltJxg3 0-1
not look too bad for White, who can A fine positional game from the
use the d6-outpost. However, some French Defence stalwart.
1 31
The F re n c h Ta rra s c h
Summary
As long as Black knows his stuff and doesn't get confused by any move order
tricks, he should have nothing to fear from 4 tbgf3 . Black should avoid 4 ...cxd4 5
lbxd4 lbf6 6 exd5 'ii'xd5?!, but alhhe other lines discussed in this chapter look
quite playable.
5 ...lbxd5 Game 76
-
4 ... a6
5 exd5 exd5 - Games 68-70
5 dxc5 Game 77
-
4 ...lbc6
5 exd5 exdS - Chapter 7
5 i.b5 (D)
5 ...cxd4 6 lbxd4 Game 75 (by transposition)
-
5 ...dxe4 Game 78
-
5 lllxd4 (D)
5 exd5 'iifxdS - Chapters 5-6
5 ...lllf6
5 ... lbc6 Game 75
-
6 exd5 - Game 74
132
CHAPTER TEN I
Third Move Alternatives
for B lack
1 33
The F re n c h Tarra s c h
134
Third M o ve A l terna tives for B l a c k
1 35
The F re n c h Ta rra s c h
The point to White's previous play. ing a useful square for the bishop on
Now 10 ... i.xg5 1 1 tbc6 traps the c8.
queen, so Black is forced into contor 14 . . ..c5 1 5 dxc5 'ir'xc5 1 6 °ir'd2 ltJc6
tions. 1 7 ltJxc6 'ir'xc6 1 a llac 1 'ir'b6 1 9 llc3
1 0 . . . ltJba ..id7 20 l:.fc1 lltba 21 b3 h6 22 lDd4
The other try is 10 ...tbxe5 1 1 dxe5 a5 23 a4?!
i.xg5 12 tbc6 'ii'd7 13 tbxg5 and now: Around here White starts to drift
a) 13 ...'iifxc6 14 'ii' h5 'iifd7 (14 ... g6 15 slowly but surely, until the players
'ii'h6 i.d7 16 'ii'g7 0-0-0 17 tbxf7 is agree a draw in a virtually level posi
very good for White) 15 tbxh7, when tion. 23 a3 a4 24 b4!, securing a pro
Black has to deal with the menacing tected passed pawn, would have been
threat of tbf6+. preferable.
b) 13 ... h6 14 tbxf7 'iifxf7 15 0-0 0-0 23 . . . 'ir'b4 24 'ir'f4 l::tca 25 llxca+
16 'ii'd2 i.d7 17 tba5 l':.ab8 18 c4 'iiff4 ..ixca 26 h3 ..id7 27 llc7 .tea 2a
19 'iife 2 and Black's weak pawns give 'ir'e3 llba 29 f4?! 'ir'a3 30 'ir'c3 'ir'a2
White the advantage, as in Yudasin 3 1 :ca 1h -1h
Drasko, Tbilisi 1987.
1 1 ..ixe7 'ir'xe7 1 2 c3 0-0 1 3 0-0 Game 81
cxd4 Spassky-Drasko
White also holds the advantage after Sarajevo 1986
13 ... c4 14 b4 cxb3 15 'iifx b3!, as 15 .. . c5
runs into 16 'ii'a3 ! 1 e4 e6 2 d4 d5 3 ltJd2 ltJc6 4 lDgf3
1 4 cxd4 lDf6 5 e5 ltJd7 6 lDb3
This is the theoretical recommenda
tion for an advantage against 3 ...tbc6.
White lets the c 1-bishop out and pre
pares for action against Black's ... c7-c5.
6 . . . a5 7 a4 b6 a ..if4 ..ie7 9 h4 ..ib7
10 c3 'ir'ca 1 1 ..ib5 ltJda
136
Third Move A lt e rn a t ives fo r Black
4 . . . c5 5 dxc5 ltJd7!
With the bishop already having
moved to e7, this is the most favour
able way of recapturing on cS. We
now reach a position similar to those
24 <it>h2? that arise after 3 ... cS 4 exdS exdS 5
24 'it>f2 is more accurate. The white tbgf3 tbf6 (Games 64-67), the only
king may find itself surprisingly ex difference being that the light-squared
posed on the h-file. bishops remain on the board. In any
24 . . . g5 25 hxg6 h5! case, Black's position looks perfectly
Now 26 'ii'xhS tbxg6 27 'it>gl 'it>f7 al playable.
lows Black to take over operations on 6 exd5 exd5 7 lDb3 ltJxc5 8 lDf3 lDf6
the kingside, so Spassky bailed out 9 0-0 0-0 1 0 i..e3 ltJce4!
with a timely draw offer. 10 . ..tbxd3 1 1 'iifxd3 nets a bishop
26 lDc1 % -% for a knight, but helps White to co
ordinate his remaining pieces. As is
Game 82 normal in these IQP positions, it is
Adams·Lputian generally useful for Black to keep the
Manila Interzonal 1990 pieces on.
1 1 i..e 2 i..d6 1 2 1:%.e 1 1:%.eS 1 3 i..d4
1 e4 e6 2 d4 d5 3 ltJd2 i..e 7 lDh5?!
Here Black starts to go somewhat
see following diagram
astray, weakening his position with
4 i..d 3 out any provocation. The simple
One of the ideas behind the 3 ... ii.el 13 .. . ilfS would have been perfectly
line is seen after the normal 4 lbgf3. acceptable.
137
The F re n c h Tarra s c h
Game 83
1 4 g 3 f 5 1 5 1Lf 1 1Le6?! 16 1Le5! 011-Topalov
Adams is at his best in these type of Zaragoza 1992
positions. Here he offers an exchange
of dark-squared bishops, which would 1 e4 e6 2 d4 d5 3 lDd2 a6
emphasise Black's newly formed Here is a quick look at two rare
weaknesses. moves:
1 6 . . .lDhf6 1 7 °il'd4 1Le7 1 8 h3 llf8 a) 3 .. .fS has been played from time
1 9 1Lg2 "il'e8 20 °il'd3 ltJd7 2 1 1Lf4 to time, but it looks rather ugly and
ltJdc5 22 ltJxc5 1Lxc5 23 lle2 °il'a4? gives White a nice juicy outpost on eS.
24 ltJg5! "il'c6 25 ltJxe6 "il'xe6 26 The game Karpov-Enevoldsen, Skopje
�h2 °il'f7 Olympiad 1972, which went 4 exfS
Visually Black's position looks exfS 5 lbdf3 lbf6 6 i.gS i.e7 7 i.d3
okay, but in fact it is about to fall lbe4 8 i.xe7 'ii'xe7 9 lbe2 'ii'b4+ 1 0 c3
apart at the seams. Note that 26 ...i.xf2 'ii'xb2 11 0-0 0-0 12 c4 dxc4 13 i.xc4+
27 l::!.xf2 lbxf2 28 i.xdS wins for 'it>h8 14 :b1 'iifa3 15 lbeS with a clear
White. advantage to White, has put most
27 f3 lDd6 people off this line.
b) 3 ...lbd7 4 lbgf3 lbe7!? is an un
usual idea of Petrosian. Now after 5 c3
cS 6 eS lbc6 we have suddenly reached
a main line 3 ...lbf6, with White's
king's knight committed to f3. Instead
White can consider keeping the ten
sion with 5 g3 cS 6 i.g2 cxd4 7 lbxd4,
or 5 i.d3 cS 6 c3, both of which look
slightly better for the first player.
After 3 ... a6 most White players con
tinue with 4 lbgf3 cS, transposing to
lines considered in Chapters 8 and 9.
28 llae 1 Here we look at another possibility.
138
Third M o ve A lt e rn a tives for B l a c k
Game 84
Shirov-Hi.ibner
Munich 1993
139
Th e F re n c h Ta rra s c h
140
Th ird M o ve A lt e rn a tives for B l a c k
14 1
The Fre n c h Ta rra sc h
Summary
Out of all of Black's third move alternatives, 3 ... .5le7 looks to be the most under
rated and it is surprising that it is not seen more often. After this move Black
often reaches uncharted positions similar to those that arise after 3 ...lbf6 and
3 ... cS. On the other hand, 3 ... lbc6 should cause well prepared White players no
problems whatsoever.
1 e4 e6 2 d4 d5 3 tt:Jd2
3 ...tt:Jc6
3 ... .5le7 - Game 82
3 ... a6 (D)
4 lbgf3 cS
5 exdS exdS Games 68-70
-
5 dxcS Game 77
-
4 eS - Game 83
3 ... dxe4 4 lbxe4 (D)
4 ....5ld7 Game 84
-
4 ...lbd7 - Game 85
4 tt:Jgf3
4 c3 - Game 79
4 . tt:Jf6 5 e5 tt:Jd7 (D) 6
. . tt:Jb3
6 .tbs Game 80
-
4 ... a5 Game 81
-
3 a6
. . .
4 ttJxe4 5 ltJd7
. . .
142
INDEX OF GAMES I
143
The Fre n c h Tarra s c h
144