"The Complete Dragon" by Eduard Gufeld and Oleg Stetsko, 1997
ICE American Batsford, Figurine Algebraic Notation, Softcover,
352pp., $26.95

In the Introduction to "The Complete Dragon," authors Gufeld and
Stetsko note that "Almost half a century of analysis and practical
experience by several generations of chess players, which has been
especially intense in the second half of the twentieth century, has
created a monumental body of theory." Given such a large, existing
body of knowledge and the double-edged play that often results
from the opening, the Sicilian Dragon is a variation for which the
importance of theory cannot be underestimated.

Providing comprehensive coverage of an opening such as the
Dragon is no small feat. As expected, G & S's work is a fair-sized
tome, about one inch thick, consisting of 352 pages including 78
complete games with light annotations, plus a detailed, four-page
Index of Variations. The complete games are mostly from the
1970s into the 90s, but none after 1995.

Coverage of the opening begins on move 6, after 1 e4 c5 2 Nf3 d6
3 d4 cd4 4 Nd4 Nf6 5 Nc3 g6. Variations after 1 e4 c5 Nf3 Nc6,
such as the Accelerated Dragon (3 d4 cd4 4 Nd4 g6) and the
Maroczy Bind (4 c4), are excluded.

The book is segregated into two parts and 11 chapters of material
that correspond to the 1997 edition of "Encyclopedia of Chess
Openings" Volume B (ECO), variations B70 through B79. Part 1 is
devoted to the Yugoslav Attack, probably the sharpest line in the
Dragon, during which White and Black normally castle on opposite
sides and proceed to go after each other's King. After 1 e4 d5 2
Nf3 d6 3 d4 cd4 4 Nd4 Nf6 5 Nc3 g6 6 Be3 Bg7 7 f3 0-0 8 Qd2
Nc6, Chapter 1 looks at 9 0-0-0 and Black responses such as 9 d5
and 9 Bd7, corresponding to ECO line B76; Chapter 2 focuses on
9 Bc4, ECO lines B77, B78 and B79; 9 g4, ECO B76 is Chapter 3;
Black 7th move deviations are the subject of Chapter 4 (ECO B76).
Part 2, Classical and Other Variations, focuses primarily on non-
Yugoslav lines of play. Thus, after the moves 1 e4 c5 2 Nf3 d6 3
d4 cd4 4 Nd4 Nf6 5 Nc3 g6 6 Be2 Bg7, Chapter 5 examines the
Classical system with 7 Be3 Nc6, corresponding to ECO variations
B73 and B74; more of the Classical, with 7 Nb3, ECO B74, can be
found in Chapter 6; the Levenfish line, 6 f4, ECO B71, is featured
in Chapter 7; Chapter 8 examines the system of play with 6 h3 Bg7
7 Be3 0-0 8 Bc4 Nc6, line B72 in ECO ; Chapter 9 reverts to
additional Black deviations on move 7 in the Yugoslav, B75 and
B76 of ECO; 6 Bg5 is considered in Chapter 10 (ECO line B70);
the fianchetto line, 6 g3, is the subject of Chapter 11 (also line B70
in ECO.) 

To judge completeness and accuracy of analysis, several lines were
selected based upon popular praxis as given in the "Trends"
pamphlet on the Sicilian Dragon (1993 edition), and a comparison
was made with ECO. Thus, in the Yugoslav Attack after the moves
6 Be3 Bg7 7 f3 0-0 8 Qd2 Nc6 9 Bc4 Bd7 10 0-0-0 Rc8 11 Bb3
Ne5 12 h4 Nc4 13 Bxc4 Rxc4 14 h5 Nxh5 15 g4 Nf6, Gufeld and
Stetsko offer no less than 10 White options on move 16, while
ECO offers 8. The move of choice in "Trends," at least, is 16 Nb3.
The latest word in this line, based upon two 1991 games by
Khalifman, would seem to be 16 Re8 17 Bh6 (which is
misidentified as 17 Bg5, under Variation 8Gb on page 75). ECO
follows Nunn-Khalifman, which eventually led to a draw after
17 Bh8 18 Bg5 Qc8 19 Qh2 Be6 20 Rd3 h5, while G & S deviate
with 19 Rh4 Ng4 played in Short-Khalifman, and continue with
Khalifman's suggestion of 20 fg4 Bg4 21 Re1 h5 "where Black has
enough pawns for the piece." (Chapter 2, p. 79) For the record, the
same note attributed to Khalifman is also contained in ECO.

Looking at the Classical after 6 Be3 Bg7 7 Be2 Nc6 8 0-0 0-0 9
Nb3 Be6 10 f4, Trends indicates that 10 Qc8 11 Kh1 Rd8 12 Bg1
is one of the more popular variations played at the GM level. This
line has appeared in at least 4 GM games in the 1992-1993 time
period: Anand-Gelfand, Linares 1992; Barua-Tiviakov, Tilburg
1992; Illescas-Tukmakov, Leon 1992 and Lau-Tukmakov,
Antwerp 1993. While the line is covered by both sources, ECO and
G & S, their assessments of it differ. For instance, G & S follow
Illescas-Tukmakov after 12 d5 13 e5 Ne4 14 Bd3 f6 15 ef6 ef6
16 Nb5 f5 17 c3 Kh8, won by White, while ECO continues with
17 Bf7 18 a4 a6 19 N5d4 Re8 20 a5 Re7 and consider the
position to be unclear, as in Lau-Tukmakov. G & S include Lau-
Tukmakov in a note on page 191 and comment that "the rook
breaks into the queenside" with, presumably, the advantage, if one
continues to follow Lau-Tukmakov: 21 Ra4 Be8 22 Nc6 Bc6 23
Rb4. 

In an opening where the play is often balanced on a razor's edge
such as in the Dragon, any new move could potentially change the
evaluation of an entire line and, suddenly, turn a previously unclear
position into a significant advantage for one side or the other.
Current theory is, therefore, of paramount importance to the
Dragon practitioner. Of course, G & S include all of the Dragon
games from the 1995 Kasparov-Anand PCA World Championship,
the games which formally inaugurated the Dragon into the annals
of World Championship play. All are included in chapter notes,
with several provided in their entirety in the Illustrative Games
section. 

Reviewing Game #17 analysis by G & S (page 106 and Illustrative
Game #17) and Anand, who annotated the game in "New In
Chess" magazine (NIC) 1995 issue #7, after 6 Be3 Bg7 7 f3 0-0 8
Qd2 Nc6 9 Bc4 Bd7 10 h4 h5 11 0-0-0 Rc8 12 Bb3 Ne5 13 Bg5
Rc5 14 Kb1 Re8 15 Rhe1 Qa4 16 a3 b5, Black's last move is
questioned by both Anand and G & S, with the latter calling it a
"blank shot" on page 106 and recommending 16...Nc4 17 Bc4 Rc4
as "more natural and stronger" in the Illustrative Games.
Continuing, after 17 Bf6 ef6 18 Nde2 Rc6 19 Nd5 (G & S give it a
"!" while Anand likes Speelman's idea of 19 Qf4, instead) 19...Qd2
20 Rd2 Nc4 21 Bc4 bc4 22 Red1 f5 23 ef5, G & S mention on
Page 106 that "23 Ng3!? seems stronger and then Black must work
harder for a draw" and cite 23 Ng3 again on page 304, attributing it
to Wahls, but offer no further analysis. Although Anand remains
silent on this move in NIC, G & S state that Anand doesn't
consider 23 Ng3 to be "good enough" (p. 304). The game was
eventually drawn.

A lesser-known game omitted by G & S, but which may have some
potential theoretical implications, is Raaste-A. Fedorov, St.
Petersburg 1996, analysed by Gennady Nesis in NIC 1997 issue #1.
In the Yugoslav Attack after 6 Be3 Bg7 7 f3 0-0 8 Qd2 Nc6 9 0-0-
0 Be6 10 Kb1 Rc8, G & S follow Liss-S. Farago, Budapest 1995,
which saw White obtain the advantage after 11 Ne6 fe6 12 Bc4
Qd7 13 Bb3 b6 14 Bh6 Na5 15 h4 Nb3 16 cb3 Rf7 17 Bg7 Rg7 18
e5. (Page 46) Raaste-A. Fedorov, won by Black in 30 moves, saw a
different approach: 11 Nb3 a5 12 a3 Ne5 13 Nd5 a4 14 Nf6 Bf6 15
Nd4 Nc4 16 Bc4 Bc4 17 h4 d5!. The answer may lie in the fact that
11 Nb3 is simply inferior to 11 Ne6 but, nonetheless, this writer
would've liked to have seen the game included.

Other relatively recent material found in NIC (1996 issue #1) and
contained in G & S includes Topalov's quick win with Black
against Ivanchuk from Belgrade 1995: 6 Be3 Bg7 7 f3 0-0 8 Qd2
Nc6 9 Bc4 Bd7 10 0-0-0 Rc8 11 Bb3 Ne5 12 h4 h5 13 Bg5 Rc5 14
g4 hg4 15 f4 Nc4 16 Qe2 Qc8 17 Bf6 Bf6 18 Nd5 Rd5 19 ed5 b5
20 h5 g5 21 fg5 Bg5+ 22 Kb1 f5 23 Rd3 f4 24 Bc4 Qc4 1-0 (pp.
96-7)

Of note is Gufeld's article in NIC 1997 issue #3 entitled "Dragon
Variation Counter Fianchetto" which, essentially, offers no more
than a reiteration of Chapter 11 of The Complete Dragon (6 g3).
Material is nearly identical, with the exception of changes to some
of the introductory narrative.

Although there does not appear to be much new in the last year or
two in the Dragon, kudos to G & S for having capturing all, or
most, of the current material available. Certainly the authors cannot
be faulted for the lack of GM and IM tournament praxis.

It must be mentioned, however, that an index of players/games
would have been convenient. For example, in searching for the
Anand-Kasparov games from the 1995 PCA World Championship,
this writer was forced to first find them in another source, and then
locate them by specific move order in The Complete Dragon, a
tedious process that should be unnecessary. An index of Illustrative
Games would have cured this ill.

Gufeld and Stetsko have put together a definitive work on one of
today's most heavily analyzed openings. This is heavy duty stuff,
however, probably best left in the hands of the experienced
Dragoneer to use as a reference source for his favorite opening. Of
course, if you ARE (or anticipate being) a Dragon practitioner (or a
Dragon slayer), clearly this is a book you want in your library,
because you can bet it will sit prominently on the bookshelf of your
opponent.