The Kibitzer 
by Tim Harding 

Transpositions

A FEW WEEKS AGO, I received a reader enquiry asking me to
explain transpositions in the opening. At first I thought this was a
beginner's question, easily answered in two minutes by email, but I
soon realised the topic was far more complex and needed a whole
article to cover it properly. However, I could not deal it with at the
time because I was committed to writing two columns about the
New York 1924 tournament. However, better late than never!

Let's start with a simple definition. In chess, especially opening
theory, the term "transposition" refers to the reaching of a position
by one move order that normally comes about from a different
sequence. Sometimes it is just because one player plays two moves
in reverse order but sometimes it can become more complicated.
There can be various reasons for transpositions and the effect can
be anything from trivial to quite startling. 

The word "transposition" can also occasionally arise in cases of
middle-game or endgame analysis but this article won't cover that.
In that case the term means that a position arising in one variation
of analysis is identical to one in another variation. Chess computer
programs use an indexing device called "hash tables" to cope with
this and prevent them analysing the same position twice; they need
these because they are really quite stupid and spend a long time
looking at moves which are quite meaningless and which even the
average human club player does not waste time analysing.

I think the topic of transpositions in the opening is large enough for
one column.

So let's start with a simple case and move to more complex
examples later. 1 e4 e6 2 d4 d5 is the standard way to arrive at the
French Defence. 

Say you like to play this defence a lot, so you answer 1 d4 by 1...e6
also. If your opponent obliges by replying 2 e4 then you have a
French Defence. This is about the simplest possible example of a
transposition. Of course your opponent will probably not play 2 e4
unless his main reason for not beginning 1 e4 is to avoid, for
example, the Sicilian. More likely, White has opened 1 d4 to get a
closed opening and the next move will be 2 c4 or maybe 2 Nf3.
Your attempt to get a French will be futile 95% of the time.
However if you also play the Classical Dutch Defence (i.e. 1 d4 f5
followed by ...e6) but don't like the Staunton Gambit (1 d4 f5 2 e4)
then you can play 1 d4 e6 as a mini-repertoire. You are then
equally happy if your opponent allows 1 d4 e6 2 e4 (French
Defence) or 2 c4/2 Nf3 f5 (Dutch).

There is another way to reach the above diagram position1 d4 d5 2
e4!? e6. White offers the Blackmar Diemer Gambit and Black says,
no thanks, I'll have a French. Some openings cannot even be
played without offering the opponent the chance to transpose to
different ones. BDG players know this. Apart from 2...dxe4 and
2...e6, they often meet 2...c6 (a Caro-Kann by transposition). It
soon gets more complicated than that, without going beyond move
2 or 3. You like to play the Scandinavian Defence (Center
Counter), so you meet 1 e4 by 1...d5. White doesn't have to play 2
exd5; he can offer a BDG by 2 d4 and then you have the choice
above - accept, play a French or play the Caro-Kann.

White can also meet the Scandinavian with 2 Nc3, with a position
that can also arise via 1 Nc3 d5 2 e4. Some "irregular" openings,
like 1 d3 and 1 e3, are primarily important for the transpositional
possibilities to which they give rise. The opening 1 Nc3, which
was the subject of Kibitzer 25 last year, is in-between - there are a
lot of transpositional lines, but also there are quite a few variations
which are distinctive of 1 Nc3.

Here is an example of how 1 Nc3 can be effective as a
transpositional ploy. Many years ago I was playing in a tournament
where my most dangerous rival employed the Sicilian. So instead
of opening 1 e4 against him, I played 1 Nc3 and he incautiously
answered 1...e5; now I played 2 e4 with a Vienna. He wasn't at
home in that opening and I won.

Here is another opening that you cannot insist upon - From's
Gambit. If you play Bird's Opening, 1 f4, you have two choices -
learn the accepted From (1 f4 e5 2 fxe5) or play the King's Gambit
(1 e4 e5 2 f4). Look at the position after 1 f4 e5.

Many Bird players select 2 e4!? in this position, reaching the
King's Gambit by simply transposing the first and second moves.
You don't have to learn any other 1 e4 openings, just the King's
Gambit in all its forms and then you play the Bird without knowing
anything about the From. It's a question of style (do you prefer to
attack with material down or defend with material up?) and
analysis: both the From (for Black) and King's Gambit (for White)
could be unsound but both these are open questions. From Gambit
theory is less well known but it is also more likely to have mistakes
as it is based on far fewer grandmaster games, and you also have to
reckon that if your opponent replies 1...e5 to 1 f4, he may also do
so in "real life" (i.e. against 1 e4) and will have had much less
practical experience with 2 fxe5. My advice to Black, having lost a
game with it, is don't play the From; my advice to White is play 2
fxe5 or better still don't play 1 f4.

Transpositions in the very early moves of a game are generally
trivial if they involve one player reversing the order of two moves
while his opponent's moves don't alter. For example, it's easy to
see that 1 e4 e5 2 Nc3 Nf6 3 Bc4 and 1 e4 e5 2 Bc4 Nf6 3 Nc3
give the same position, which some writers classify as a Vienna
Game and others as a Bishop's Opening. Similarly it is not
uncommon for games with the Nimzo-Indian, King's Indian and
other 1 d4 Nf6 defences to arise from 1 c4 Nf6 2 d4. This is quite
common in master play.

A GM may like certain variations of an opening but want to avoid
others, or may want to invite certain defences but prohibit others.
For example, White can invite the King's Indian but deny the
Gruenfeld by playing 1 c4 Nf6 2 Nc3 g6 3 e4 and if 3...Bg7 then 4
d4. Black can play differently, e.g. 3...e5, but White may prefer this
if he doesn't like Gruenfelds.

So much for very simple transpositions that anybody can spot.
Let's move on to slightly more complicated examples. I will list a
few cases where a well-known opening position can arise by two
move orders, beginning with some examples where the
transpositions are only between different variations of the same
opening, then we will move to cases where a different opening is
reached, but at a later stage than in the trivial cases above.

Case A. French Defence 1 e4 e6 2 d4 d5 3 Nc3 Nf6 4 e5 Ne4!? 5
Nxe4 dxe4.

This can also arise via 1 e4 e6 2 d4 d5 3 Nd2 Nf6 4 e5 Ne4 5
Nxe4, etc., but note that in this move order Black is not threatening
to double the c-pawn so White can also consider 5 Bd3. This
illustrates the point that just because a position may be playable via
one route, there can be problems (or at least more to learn) if you
try to reach it even when the opponent plays a slightly different
move.

Case B. Slav Defence. 1 d4 d5 2 c4 c6 3 cxd5 cxd5 4 Nc3 e5!?.
The Winawer Counter-Gambit. 

This normally arises via 1 d4 d5 2 c4 c6 3 Nc3 e5!? when 4 cxd5
cxd5 is just one line, and White has the better option 4 cxd5 d4 5
Ne4 Qa5+ 6 Bd2. However in the Exchange Variation order with 3
cxd5 cxd5, White sometimes plays 4 Nc3 (instead of the regular 4
Nf3) with the idea of meeting 4...Nc6 by the lively 5 e4!?, but
Black gets his say first by 4...e5!? and in this case the gambit is a
better bet because White has exchanged c-pawns.

Case C. Sicilian Defence, Sveshnikov Variation. The diagram
shows the standard position reached after 1 e4 c5 2 Nf3 Nc6 3 d4
cxd4 4 Nxd4 Nf6 5 Nc3 e5 6 Ndb5 d6 7 Bg5 a6 8 Na3 b5.

However, Adorjan and Horvath in their 1987 book on the
Sveshnikov, gave as their standard move order1 e4 c5 2 Nf3 Nc6 3
d4 cxd4 4 Nxd4 Nf6 5 Nc3 e6 6 Ndb5 d6 7 Bf4 e5 8 Bg5 a6 9 Na3
b5. The first thing to note is that the position is indeed the same but
each player has "lost" a move: White by playing Bc1-f4-g5 and
Black by playing ...e6-e5. So the move number is different by one;
this is not the only well known example in chess theory where this
can happen.

The Hungarian masters explain in their introduction why they
prefer this move order, after 5...e6, Black has really only one
alternative variation to learn (6 Nxc6) whereas in the move order
with 5...e5 they point out "Black has to consider six possible knight
moves, and after the strongest 6 Ndb5 d6, apart from 7 Bg5, White
also has the possibilities of 7 a4 and 7 Nd5".

Case D. Scotch/Two Knights Defence1 e4 e5 2 Nf3 Nc6 3 Bc4 Nf6
4 d4 exd4. 

This well-known variation can arise via various move orders, e.g. 1
e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Bc4 Nf6 or the Petroff (1 e4 e5 2 Nf3
Nf6 3 d4 exd4 4 Bc4 Nc6), or the Bishop's Opening (1 e4 e5 2 Bc4
Nf6 3 d4 exd4 4 Nf3 Nc6) or even via the Centre Game, 1 e4 e5 2
d4!? exd4 3 Nf3 Nc6 4 Bc4 Nf6.

Case E. Petroff to French Exchange. 1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6
4 Nf3 Nxe4 5 d3 Nf6 6 d4 d5. 

This actually occurred in one of my games. Although Black's sixth
move looks sensible, taking an equal share of the centre, he should
maybe avoid symmetry by 6...Be7 or some other move. After 6...d5
the position is identical to the Exchange French line, 1 e4 e6 2 d4
d5 3 exd5 exd5 4 Nf3 Nf6. Note that in this case, the move number
differs from the norm by two! The reason for this is not the double
moves of the pawns (they cancel out) but because both players
made three moves with their knights to get them back to a square
reached after their first move.

Case F. Semi-Slav to Caro-Kann or French1 d4 d5 2 c4 e6 3 Nc3
c6 4 e4!? Bb4 (Black declines the dangerous Marshall Gambit.) 5
exd5 and now 5...cxd5 6 Nf3 Nf6 is a Caro-Kann Panov Attack, in
fact a variation of it with Nimzo-Indian affinities! Black can also
play 5...exd5 with a French Defence Exchange sort of position. In
that case, the position may not be one that would normally arise via
the French, though it is possible (1 e4 e6 2 d4 d5 3 exd5 exd5 4 c4
c6 5 Nc3 Bb4) but the main point is the character of the position.
White wanted a wild gambit and has instead been lured into a
symmetrical position where he may end up with a weak d-pawn.
Probably White should therefore prefer 5 e5 or some other move.

Case G. Alekhine's to French Defence1 e4 Nf6 2 Nc3 d5 3 e5;
White avoids the main lines of the Alekhine's Defence and tests
his opponent's knowledge of the lesser variations.

Both 3...d4 and 3...Ne4 are playable for Black here. However, one
of my opponents played 3...Nfd7 which sets a trap because 4 Nxd5
is not good for White. However, I played 4 f4! and Black had to
defend a main line Steinitz French: 4...e6 5 d4 c5 6 Nf3 Nc6 7 Be3.

Case H. Now we are getting to more advanced examples. This
occurred in one of my own postal games and was already discussed
in detail in my book "Winning At Correspondence Chess", but I
bring it up again here as it is one of the best and strangest examples
of transpositions from one opening to another that I know. After 1
e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 e6 6 Nf3 we have a
Panov-Botvinnik variation of the Caro-Kann (ECO classification
B14), but this line is particularly significant as it leads to probably
the most common case where a 1 e4 opening transposes to a 1 d4
opening. This happens if Black chooses 6...Be7 7 cxd5 Nxd5 8 Bc4
Nc6 when we have a Queen's Gambit Semi-Tarrasch (D41). 

This position can arise (but in practice does so more rarely) via 1
d4 d5 2 c4 e6 3 Nc3 Nf6 4 Nc3 c5 5 cxd5 Nxd5 6 e3 Nc6 7 Bc4
cxd4 8 exd4 Be7. What happened next in my game made the
example more strange and instructive. Play continued 9 0-0 0-0 10
Re1 Nf6 11 a3 a6 (see next diagram) and now we had a variation
of the Nimzo-Indian (E57) although Black never put his bishop on
b4! So three different volumes of "ECO" were needed to track the
opening of this game. It is well known that Karpov's favourite
move 6...Bb4 (instead of 6...Be7) can lead to Nimzo-Indian
positions, but the transposition in my game was far from obvious
as Black played 6...Be7. 

So how can this be a Nimzo when the black bishop never went to
b4? Well, the position after my 11th move can also arise via 1 d4
Nf6 2 c4 e6 3 Nc3 Bb4 4 e3 0-0 5 Bd3 d5 6 Nf3 c5 7 0-0 Nc6 8 a3
dxc4 9 Bxc4 cxd4 10 exd4 Be7 11 Re1 and that is how opening
books deal with it. I spotted the transposition and found a good
idea for White a few moves later in Tony Kosten's book on the
Nimzo-Indian. My opponent either didn't have that book or
thought we were still in a Caro-Kann, or a Semi-Tarrasch, and
missed my winning move.

As you can imagine, there are not many other examples where 1 e4
openings transpose to 1 closed openings or vice versa (except for
cases like 1 e4 g6 2 d4 Bg7 3 c4) but there are some lines in the
Sicilian where White plays c2-c4 that have close affinities with the
Symmetrical English.

In some openings, the possible transpositions form a maze that can
be quite baffling for the learner, and even the authors of
monographs on the openings can be caught out. Attentive readers
of my own books can maybe point out one or two cases where I
have made such mistakes too! You cannot expect authors to point
out every possible transposition, especially in openings where the
play is non-forcing and different sequences are common, but
unusual ways of reaching identical positions do sometimes call for
comment.

Particularly annoying for readers, of course, are cases where a
position is analysed in two different chapters or sub-sections of the
same book, but with differing examples and assessments! As they
say in the army, "no names, no pack drill" so no individuals will be
castigated, but later in this column you will find an example where
the famous "Encyclopaedia of Chess Openings" has made a
mistake of this kind.

Move order tricks and variants are a major part of many
professional players' armory, especially in the closed (i.e. non-1
e4) openings that give greater opportunity for this . Playing at a
level where all their opponents' games and their own games are
published as soon as they are played, grandmasters constantly seek
nuances to steer for their favourite lines while avoiding ones that
their opponents favour. They try to get opponents into lines that are
inferior or little-known, or if they cannot do that then at least they
try to avoid the lines they think the opponent most wants.

Look for example at a case from last year's Shirov-Kramnik match
in which Shirov emerged as surprise winner for two basic reasons:
he avoided the snares of Kramnik's very deep opening preparation
and when the opening was finally over, Shirov played more
creatively. I am particularly thinking of game seven. Exchange
Gruenfelds in Games 1, 3 and 5 had all ended in draws and Shirov
won game 4. In game 7, after 1 d4 Nf6 2 c4 g6 Kramnik switched
from 3 Nc3 to 3 g3, probably hoping for a King's Indian or an
exchange variation of the fianchetto Gruenfeld. Shirov ruled all
that out by 3...c6!, only playing ...d5 a couple of moves later; he
drew again and maintained his lead.

Avoiding lines you don't like or which are theoretically suspect is
very important in chess. Another example comes from "The
Complete Benoni" by GM Lev Psakhis (Batsford 1995) in which he
admitted that he gave up playing the Modern Benoni via the
standard move order 1 d4 Nf6 2 c4 c5 3 d5 etc. The reason is that
White was having a lot of success with the line 4 Nc3 exd5 5 cxd5
d6 6 e4 g6 7 f4 Bg7 8 Bb5+; a famous example of that line is
Kasparov-Nunn, Lucerne Olympiad 1982. 

That game went1 d4 Nf6 2 c4 e6 3 Nc3 c5 (Note that Nunn has
also played 2...e6, probably trying to avoid some other line he
didn't like in the Benoni, but in this case White has developed the
queen's knight. The punishment was swift) 4 d5 exd5 5 cxd5 d6 6
e4 g6 7 f4 Bg7 8 Bb5+ Nfd7 9 a4 Na6 10 Nf3 Nb4 11 0-0 a6 12
Bxd7+ Bxd7 13 f5 0-0 14 Bg5 f6 15 Bf4 gxf5 16 Bxd6 Bxa4 17
Rxa4 Qxd6 18 Nh4 fxe4 19 Nf5 Qd7 20 Nxe4 Kh8 21 Nxc5 1-0.

Therefore, Psakhis reasoned that Black was OK in most Benoni
lines, he should wait for White's 3 Nf3 before committing to ...c5.
So after 1 d4 Nf6 2 c4 he played 2...e6 and if 3 Nc3 he could play a
Nimzo-Indian or some form of Queen's Gambit, but if 3 Nf3 then
he would play 3...c5. Now, unless White wants a Symmetrical
English (4 c4) he plays 4 d5 perhaps expecting the suspect
Blumenfeld Counter-Gambit (4...b5?! 5 Bg5!). Instead Black can
play 4...d6 5 Nc3 g6 with a regular Modern Benoni.

Reverse transpositions are another issue: sometimes it is possible
to reach a well-known position with reverse colours and an extra
move. For example, the move 1 a3 could be played with this
objective and in the English Opening there is a variation which is
really a reversed Dragon Sicilian. These cases are not quite the
same thing as a regular transposition and really form a separate
topic.

Anybody who plays chess regularly gradually develops an opening
repertoire, whether haphazardly (like most amateurs) or a carefully
constructed maze of interconnected variations (grandmasters). This
repertoire is built out of positions that you like to play   positions
where you feel at home and confident   but those positions have
to be reached, so in building a repertoire you have to consider both
your preferred variations and the best ways to reach them. 

If a problem arises with the main move order to your favourite
variation, you can try to find a transposition, an alternative and
safer sequence to the same position. Every move you make is a
process, but the position thereby arrived at is what the opponent
has to deal with. A few rare moves are startling enough in
themselves to have psychological effects on the opponent but it is
the position arising on the board that really matters. One way to
play against the Sicilian is the "chameleon" variation, 1 e4 c5 2
Ne2, which really disguises your intentions. Against most replies,
White has the choice of a closed formation or transposing to
regular open variations with a subsequent d4. (See Diagramk)

The effectiveness of 2 Ne2 depends partly on knowing which
Sicilian variation your opponent prefers as this will help you
decide whether or not to play d2-d4 (usually after 3 Nbc3 to keep
Black guessing a bit longer). It is particularly effective against
Najdorf fans because an early ...d6 rules out some of Black's best
options in a closed Sicilian.

The more forcing the play in the early stages of a game, i.e. the
more the two players' forces interact to create direct threats, the
less likely it is that you will be able to find a suitable transposition.
This is why transpositions and move-order tricks occur much more
often in the closed openings where White makes a first move other
than 1 e4, and especially in the Reti and English openings. 

Even in a sharp opening like the King's Gambit, however,
transpositions can occur. Fischer's Defence to 1 e4 e5 2 f4 exf4 3
Nf3 d6, now one of the main lines of the gambit but virtually
unheard of until the 1960s, was designed as a way for Black to
reach the more promising of the old classical 3 Nf3 g5 lines (e.g.
the Hanstein Gambit) while avoiding the riskier ones like the
Kieseritzky (1 e4 e5 2 f4 exf4 3 Nf5 g5 4 h4 g4 5 Ne5). White can
get a pawn structure similar to the Kieseritzky but only by
retreating his knight to g1 (3...d6 4 d4 g5 5 h4 g4 6 Ng1). So White
often avoids an early d4 against this defence to eliminate
unfavourable transpositions to some of the classical lines in the
King's Gambit.

Readers who use chess database programs will know that they
usually involve opening classification keys. For example,
ChessBase will automatically attempt to classify any game to its
ECO (Encylopaedia of Chess Openings) code, the same code that
is used by the Informator series of game collections. A whole
database of hundreds of thousands of games can be rapidly
classified by ChessBase on a fast computer. 

The correct process to classify a game accurately should be to play
through the game backwards from the end, checking after each
move retracted whether the position then arising matches a position
in the Informator opening key (with the same player to move). If it
does, then the keycode associated with that position is the ECO
classification for that game. If not, retract another move and try
again, repeating this process until a position in the classification is
reached. Assign that code and stop. It's a simple algorithm for the
computer to follow, and all it requires is that the positions in the
key are correctly assigned.

You can see how it works by inputting a transpositional variation
into your database program and seeing what classification the
position gets. For example, after 1 e4 c6 2 c4 e5 (Caro-Kann) save
the line and you should get the ECO classification B10. Now
continue with the moves 3 Nf3 d6 4 d4 Nf6 5 Nc3 and the game is
beginning to look like an Old Indian. However it is only after you
add 5...Nbd7 (or 5...Be7 6 Be2 Nbd7) that ChessBase 7.0
recognises that a position classified as A55 in ECO has arisen!
This is because the position after 5...Be7 is not included in the
database as a position linked to the key A55 but once ...c6 and
..Nbd7

So does ChessBase make mistakes, i.e. does it assign keycodes
incorrectly? Occasionally people have suggested to me that this can
happen, and I would be interested in examples any readers can put
forward. One possible cause of error is that the pieces could all be
on the same squares but it is the wrong player to move; when I ask
ChessBase to look for all games I have with a certain board
position it doesn't seem to distinguish between White or Black to
move. This could make a big difference in an ending, and can
mean in certain openings that a position which might appear to be a
transposition in fact is not.

On the whole, I find the opening classification in the program is
very quick and accurate but from time to time a game will end up
in my database under the wrong keycode. This can happen because
the person who input the game may have set the keycode
incorrectly, possibly in another program or in a PGN file that has
been imported to ChessBase, but you can always get a database
reclassified.

Nevertheless it is possible to catch ChessBase 7 in classification
mistakes, though I had to try quite a few tricky sequences before I
succeeded. The recent book "The Sicilian with ...Qb6" by Zoran
Ilic deals with a complex of lines that are bedevilled by
transpositional tricks and Ilic did a tremendous job of exploring all
the ramifications of these lines to produce a viable repertoire for
Black. I acted as Batsford's editor on this book last year and
became quite familiar with the tricks involved.

After 1 e4 c5 2 Nf3 Nc6 (he also deals with the allied 2...e6 3 d4
cxd4 4 Nxd4 Qb6) 3 d4 cxd4 4 Nxd4, Ilic recommends the move
4...Qb6. This is B32 in ECO, and the version of ChessBase on my
computer classifies that correctly. After 4...Qb6 play normally goes
5 Nb3 Nf6 6 Nc3.

This position, as Ilic correctly indicates, is to be found under B33
in ECO (and in John Nunn's "Beating The Sicilian 3") because
there it is given as arising via the move order 4...Nf6 5 Nc3 Qb6 6
Nc3; however this is an inaccurate move order because White can
play 6 Be3! instead of 6 Nb3. If you input this sequence (1 e4 c5 2
Nf3 Nc6 3 d4 cxd4 4 Nxd4 Qb6 5 Nb3 Nf6 6 Nc3) you will
probably find that ChessBase classifies it as B32 instead of B33. 

When you add the move 6...e6 and save again, the classification
switches to the correct B33.

I am not a programmer but I think there are three possible reasons
for this a) The precise position after 6 Nb3 is not included in the
electronic version of the ECO key used by ChessBase (so the game
gets classified on the basis of the position after 4...Qb6); b) The
position after 6 Nb3 is included but is assigned to the wrong
keycode; c) The program does not hunt backwards for the latest
coded position. On the basis of my usage of ChessBase, I think the
problem with this and other examples that readers may find is
either a) or b), and it is possibly a) in this case.

Continuing with the interesting examples from Ilic's book, the
author and I discovered (see pages 136-137) a position which is
given two different classifications in "ECO" depending on the
move order used to reach it. (This probably means that the position
itself is not in the key.) Not surprisingly, ChessBase makes the
same mistake as presumably they bought their copy of the ECO
key from Sahovski Informator of Belgrade. (I am not making any
propaganda points hereIlic is from Novi Sad and so it's ironic that
he has caught them out!) 

This position is normally reached by a Richter-Rauzer and it gets
its correct B60 classification in ECO and ChessBase when you
employ the sequence 1 e4 c5 2 Nf3 d6 (or 2...Nc6 with 5...d6) 3 d4
cxd4 4 Nxd4 Nf6 5 Nc3 Nc6 6 Bg5 Qb6 7 Nb3 e6 8 Bd3 Be7 9 0-0
a6 10 Kh1 0-0 11 f4 Qc7 12 Qe2. This is row 5 on page 334 of the
3rd (1997) edition of ECO. Ilic's move order to this position is 1
e4 c5 2 Nf3 Nc6 3 d4 cxd4 4 Nxd4 Qb6 5 Nb3 Nf6 6 Nc3 e6 7
Bd3 a6 8 0-0 Be7 9 Bg5 d6 10 Kh1 Qc7 11 f4 0-0. This is
discussed on pages 57-59 of his book. Now look under key B33 in
ECO and, stemming from row 3 on page 208 of the current edition,
you will see the slightly different sequence 7 Bd3 a6 8 0-0 Qc7 9
Bg5 Be7 (this is now note 17 on page 212) 10 f4 d6.

We don't have quite the exact position yet, but if you continue that
note one more move, you will see 11 Qe2 0-0 12 Kh1 and at last
we have the above diagram, exactly the same position classified as
B33! Under B33 the example given continues 12...b5 13 Rae1 h6
14 Bh4 b4 15 Nd1 Nd7 16 Bxe7 12...b5 13 Rae1 h6 14 Bh4 b4 15
Nd1 Nd7 16 Bxe7 Nxe7 17 Ne3 e5 18 Rf2 citing Kaminski-
Petrienko, Katowice 1992. Informator 56 and ECO classify this
game as B33. 

By comparison, the games cited in the relevant notes in key B60
(page 335 of ECO) are Van der Wiel-Benjamin, Wijk aan Zee
1989 (Informator 48), in which Black played 13...Nb4?! (instead of
...h6) and Wang Zili-Gostisa, Belgrade 1988 (Informator 46), in
which 13...b5-b4 was selected. All these lines are given as
favourable to White but Ilic cites other games saying "In practice
Black has achieved solid positions with the continuations 13...Bb7
and 13...Nd7 that are not mentioned in ECO..."

Evidently there are still hidden transpositions to be found in the
Sicilian and many other openings and there are points to be won by
the players who discover them.

Copyright 1999 Tim Harding. All Rights Reserved.