The Kibitzer
by Tim Harding

Interesting Byways in the Classic Open Games

The question of whether Black should meet 1 e4 by 1...e5 is one of
the toughest for players to decide and for theory to settle. The
Sicilian Defence, 1...c5, is the most popular at all levels and is
favoured both by ultra-tactical players like Kasparov and strategic
ones like Ulf Andersson. The French Defence, which was being
played by only a couple of leading grandmasters in the 1970s
(Uhlmann and Korchnoi) has had a tremendous resurgence of
popularity in the past 10-15 years, and the Caro-Kann also stands
unrefuted if you don't mind a high percentage of draws with Black.

Nevertheless there is something special about 1...e5 and it's not
merely its antiquity and the classic gambits it gives rise to. There is
a lot to be said for teaching these openings first to juniors and
beginners - provided they are taught to handle them dynamically
and not just trot out a symmetrical four knights with Bc4/...Bc5.
These openings teach a lot of the basics of threat and defence and
provide a schooling in accurate calculation which will stand
players in good stead in later years.

They are also worth revisiting by every player from time to time. In
the late 1960s I mostly played closed openings, heavily influenced
by players like Petrosian, and then postgraduate studies enforced a
two-year break from chess. When I got my masters (which at
Oxford they call a B.Phil.) I started playing chess again, and
reverted to 1 e4 and some other old-fashioned openings like the
Colle System, until I could catch up on theory. I found myself
attracted to romantic openings in a way I hadn't been in my teens.

I found two books particularly useful. I got the book on Morphy
and Paulsen in the Weltgeschichte des Schachs seriesthe ones that
have a diagram after every fifth move by Black, so you don't need a
board. I won this as a prize in my section at Wijk aan Zee in 1971
and I spent the next few weeks working through all the games in
my head; a high proportion of these were with 1 e4 e5. 

Then I bought a copy of Estrin's book on the Two Knights
Defence; this was before the English translation became available.
Probably because it was in Russian, I paid much more attention to
working through the book in detail, making lots of notes in the
margin and emphasising the bits I thought most important so I
could find them again. Then I got the Weltgeschichte des Schachs
book on Adolf Anderssen and studied the open games there too.
All this time I opened 1 e4 and met 1...e5 by 2 Bc4 which usually
gained some time on the clock as almost nobody except Bent
Larsen took that move seriously in the early 1970s.

After a while I stopped answering 1 e4 by 1...e5 as I took up the
Pirc and sometimes played Sicilians too, because of book
commissions, but I never found the Sicilian suited my style. When
the Pirc became better-known, I switched to the French, which
suited me better, but I have always thrown in a few 1...e5 games
each year "to keep my hand in"

I have also dabbled in most of the 1 e4 open games with White at
one time or another; a few of them I know very well.

So I was very interested in the recent publication of the book Play
the Open Games as Black by English GM John Emms. He deals
with all the "old-fashioned" openings that can arise after 1 e4 e5,
assuming Black meets 2 Nf3 by 2...Nc6 (he doesn't look at other
second moves for Black.).

This interesting book is subtitled 'What to do when White avoids
the Ruy Lopez', so in this article I shall not consider the Lopez
either, although of course anyone who defends 1 e4 e5 2 Nf3 Nc6
has to have something ready for 3 Bb5. At other times and in other
places, I have pointed Black in the direction of the Marshall Attack
and see no reason to do otherwise now, although I have
occasionally dabbled in other defences to 3 Bb5.

Obviously there is not space in a column of this kind to look at all
the openings covered by GM Emms in a 224-page book, so I shall
just pick out a few which interest me and which reveal some of the
strengths and weaknesses of this book. I will concentrate on lines
in which White plays Bc4, where the general policy recommended
by Emms is based on the Two Knights Defence and not playing
...Bc5 at all. 

Firstly, what about White's alternatives at move 2 after 1 e4 e5?
There are three chapters (43 pages) on the King's Gambit, which I
leave to someone else to dissect in detail. After 1 e4 e5 2 f4 exf4 3
Nf3 (two chapters are devoted to the other branches) Emms
considers just two lines: the classical 3...g5 and, interestingly,
3...h6 rather than the better-known Fischer line 3...d6. Centuries of
analysis have failed to refute 3...g5 which makes it the best choice
for correspondence play, but it does require some nerve and
memory in over-the-board and blitz play so it is understandable
that Emms wished to offer an alternative.

The Vienna Game (2 Nc3) and Bishop's Opening (2 Bc4) are
treated together in chapter 3; since several previous Kibitzer
articles have dealt with some of the lines that can arise here, I will
confine myself to a few comments.

Firstly, the Urusov Gambit (which has a lot of followers on the
Internet) is not challenged by Emms. After 1 e4 e5 2 Bc4 Nf6 3 d4
he gives us just three and a bit lines of text"3...exd4 4 Nf3 (4 e5
d5! is good for Black) 4...Nc6 gives us the Two Knights Defence
(see Chapters 15-17)".

Yes, that's all he says about it! Of course, this transposition is a
perfectly viable option for Black but it is also one that anyone who
plays 3 d4 in the Bishop's Opening is ready for. Note that this can
also arise via the Petroff Defence (1 e4 e5 2 Nf3 Nf6 3 d4 exd4 4
Bc4). It's not clear whether Emms really believes that 4...Nc6 is the
best move, or whether he is just trying to save effort and pages by
taking the easy way out, and this kind of transposition manoeuvre
is of course typical of repertoire-based opening books. "Why learn
the best move when a second-best move saves a lot of study and
memorising?" is of course quite a valid approach for amateur
players with limited time for study, but I am surprised that a GM
and professional writer doesn't at least say something about the
Urusov proper. After all, he does devote some space to moves such
as (1 e4 e5) 2 c4, 2 c3, 2 Bb5 and 2 Ne2.

Similarly, transposition to the same line is all he offers against the
Scotch Gambit, 1 e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Bc4 when 4...Nf6
gives the Two Knights. Emms does not mention other moves such
as 4...Bc5 because anything that can lead to a main line Italian
game or Evans Gambit is beyond the scope of the book,

According to a recent correspondent, the Urusov Gambit is refuted
by the following line, where Black returns the pawn to break the
attack.

1 e4 e5 2 Bc4 Nf6 3 d4 exd4 4 Nf3 Nxe4 5 Qxd4 Nf6 6 Bg5 and
now 6...Nc6 (See Diagram)

As an example of White's difficulties, the reader cites the following
master game.

K.Gneiss - Helgi Gretarsson Velden open 1996

From the last diagram, this game continued 7 Qh4 d5! (The point.
7...Be7 returns to the main line of the Urusov Gambit, as analysed
in Kibitzers 28 and 29.) 8 Bxf6 gxf6 9 Be2 f5 10 Qh5 Qf6 11 Nc3
Be6 12 0-0-0 d4 13 Nb5 0-0-0 (See Diagram)

Black has a very solid position with an extra pawn (albeit doubled
and isolated) and (more important) the bishop pair and open files
for the rooks. Gretarsson won as follows14 c3 dxc3 15 Rxd8+
Nxd8 16 Nxc3 Nc6 17 Kb1 Bg7 18 Rd1 Qe7 19 Qh4 Bf6 20 Qa4
Rg8 21 Ne1 Qc5 22 Nd3 Qb6 23 Rc1 Rxg2 24 Bf3 Rxh2 25 Nd5
Bxd5 26 Bxd5 Rxf2 27 Qa3 Rd2 28 Rxc6 bxc6 29 Qf8+ Bd8 30
Bc4 Qg1+ 31 Nc1 Qd4 32 Qb4 Bf6 33 Qf8+ Kb7 34 Ba6+ Kb6 35
Qb8+ Kxa6 36 Qc8+ Kb6 37 Qb8+ Ka6 38 Qc8+ Kb5 39 a4+ Kc4
40 Qa6+ Kd5 41 Qa5+ Ke6 0-1.

I did a little research on this and soon found that 6...Nc6 is far from
new. It was played in Tartakower-Shories, 2nd match game,
Barmen 1905 and English writer David Hooper suggested back in
the 1960s (in A Complete Defence to 1 P-K4) that this line may be
a good way for Black to avoid the complications of the gambit's
main line. So what is White to do?

After 7 Qh4 d5 8 Bxf6 gxf6 he can play 9 Bb3 instead of 9 Be2 as
in the Gretarsson game. Then 9...Be6 (9...Qe7+ 10 Kf1 Be6 11 Nc3
0-0-0) 10 Nc3 Bb4 11 0-0-0 Bxc3 12 bxc3 Qe7 13 Bxd5 0-0-0
gives about level chances as I said in my 1973 book Bishop's
Opening. However, I do not see any line after 6...Nc6 that offers
White an advantage. You could try analysing 7 Bxf6?! Nxd4 (Both
7...Bb4+ and 7...Qxf6 also need looking into.) 8 Bxd8 (8 Bxg7)
8...Nxc2+ (not 8...Nxf3+ 9 gxf3 Kxd8 10 Bxf7) 9 Kd2 Nxa1 10
Bxc7 but it is not so clear whether White can pick up the knight on
a1.

Returning to the Emms book, after 1 e4 e5 2 Nc3 Nf6 3 Bc4 he
recommends 3...Nc6. Once again, there is absolutely nothing
wrong with his recommendation but his way of dealing with the
possibility 3...Nxe4 is to pass over it in silence! Again, this is the
safest way for him and his readers, and he leaves the Frankenstein-
Dracula variation to those who have made a special study of it. 

As I mentioned earlier, if White tries to play the Urusov Gambit,
Emms recommends transposition to the Two Knights Defence, so
we get a position that can also arise via 1 e4 e5 2 Nf3 Nc6 3 Bc4
(or 3 d4 exd4 4 Bc4) 3...Nf6 4 d4 exd4. (See Diagram)

Let us look first at his coverage of the line 5 e5, which is fairly
good on the whole. He points out for example that 5...Ne4 is quite
playable as an alternative to 5...d5 because after 6 0-0 d5 7 exd6
Nxd6 8 Bd5 Nf5 9 Re1+ be7 10 Bxc6+ bxc6 11 g4 Black has
11...Nh6! as recommended in "Nunn's Chess Openings". Then if
White exchanges on h6 and swaps queens, the black bishop is
superior to the white knight as he says; I won a game with Black in
this line recently.

After 5 e5 d5 6 Bb5 Ne4 7 Nxd4 he deals briefly with the main
move 7...Bd7 but his repertoire recommendation is 7...Bc5. I just
want to look at one of the possibilities here, which reveals a
weakness in Emms' approach to research. White can play (though
it's not recommended) 8 Nxc6!? Bxf2+ 9 Kf1 reaching the next
diagram position. (See Diagram)

After 9...Qh4! this used to be thought just lost for White but in fact
it's possible White can draw. Many books give no continuation
after 9...Qh4.

Emms analyses 10 Nd4+ (his main line) to a draw and his coverage
of 10 Qxd5 is also reasonable, but he comes unstuck with 10
Nxa7+, making the comment "No mention has been made of this
move, but so far as I can see it leads to a draw..." I was totally
amazed by the first part of this comment ("no mention...") this line
was seen in print decades ago. Analysis by Dr. Hermann Kaidanz
appeared in the Wiener Schachzeitung in 1904. Subsequently
9...Qh4 was investigated by Schlechter and others, although there
have been few takers to actually play 8 Nxc6 for White.

Not surprisingly, since Emms is re-inventing the wheel in his
analysis of 10 Nxa7+, he makes several oversights. After the
forced 10...c6 11 Nxc8 Rxc8 three bishop retreats come into
consideration. Emms says 12 Ba4 draws to 12...Ng3+ but doesn't
mention Black's try for advantage, 12...Ba7 which is analysed in
note 3 on page 266 of the 8th edition of Bilguers Handbuch des
Schachspiels (edited by Schlechter); this line is also in the 7th
edition.

Emms says 12 Bd3 loses to 12...Ng3 13 Kf2 Ne4 but he doesn't
analyse the main line, 13 hxg3 Qxh1+ which also favours Black.
Again on 12 Be2 he doesn't see 12...Ba7 and gives 12...Bd4 as
leading to a draw. The continuation given in the Handbuch went 12
Be2 Ba7 13 Qe1 Nf2 14 c3 (or 14 Nd2 Qd4 15 Nf3 Qb6 16 c3
Nxh1 17 Nd4 0-0=) 14...0-0 15 Nd2 Qe7 16 Rg1 (16 Qxf2 "comes
into consideration") 16...Qxe5 17 Nf3 Ng4! 18 Qh4 (18 Qg3 Qf5)
18...Qe4 19 Ne1 (19 Rh1 Bf2 20 Qh3 Rce8 21 Nd4 f5 "and so
forth") 19...Rce8 with advantage to Black. 

As it happens, US Senior Master Mark F. Morss wrote about this
very variation in great detail last September in his own Internet
column, Hard Chess. (Highly recommendedyou can find it at
http//correspondencechess.com/campbell/hard.htm) Mark had a
game a few years ago where he actually had to meet 8 Nxc6. and
he did a tremendous amount of research into the history of the line,
none of which Emms has obviously done. He saw that Rini Kuijf, a
Dutch IM writing in New in Chess Yearbook 20, expressed distrust
of 9...Qh4 (without being aware of Kaidanz' analysis) and
recommended 9...bxc6 which was actually refuted long before
Kuijf was born.

To paraphrase Morss's article, which you should read for yourself
(it's the one entitled 'Lost Variations'), a lot of still valid old theory
has just been pruned from recent books, though it remains
unrefuted, so the only way to get some real perspective on many of
these open games is to go back and see what was known 80 or
more years ago, and what is really new.

From the last diagram, the game Clark-Morss, USCF corr 92CM76
1992, went 9...Qh4 10 Qxd5 Bc5!?

This is Kaidanz's move, of which Kuijf was unaware. Also
Beliavsky & Mikhalchishin don't have it in their new (and bad)
Batsford book on the Two Knights, although their coverage of 8
Nxc6 is somewhat more informed than that of Emms. For example,
they are the only recent source to mention the alternative retreat
10...Bb6 when 11 g3 Qh3+ 12 Ke1 Bf2+ reaching a position we
will look at in a minute, since it can also come about via 10...Bc5.
Beliavsky & Mikhalchishin also mention (10...Bc5) 11 Qd8+?
Qxd8 12 Nxd8+ Ke7! 13 Nc3 Rxd8 with advantage to Black
according to Maroczy. This is not altogether convincing as White
can improve by 13 Nxf7 or 13 Nxb7 and may be about equal. With
the bishop on c5, theory says Black can meet 11 Qd8+ Qxd8 12
Nxd8 by 12...Kxd8 but it may be that both king moves are OK in
both variationscan anyone see a significant difference?

Several writers (including me) used to say that if instead 10...Be6
11 Nd4+ Ke7 12 Nxc6 fxe6 Black was OK, but Kuijf was probably
right to say this is refuted by 13 Qd7+. After 13...Kf8 14 Qxe6
White has at least a draw say Beliavsky & Mikhalchishin;
14...Ng3+ does not work.

Probably 10...Bc5 is really the right move. (See Diagram)

Here if 11 g3 Qh3+ 12 Ke1 Bf2+ it doesn't matter where Black
played his bishop at move 10. 13 Ke2 (13 Kd1 Bg4+ 14 Be2 Qg2!)
13...Qg4+ 14 Kd3 Nc5+ 15 Kc3 and now Beliavsky &
Mikhalchishin give "15...Be6! 16 Nd4+ c6 17 Bxc6+ bxc6 18
Qxc6+ Bd7! 19 Qxa8+ Ke7 with an irresistible attack (Moller-
Maroczy, Copenhagen 1907)." Unfortunately I don't know the
conclusion of this game, but in view of 20 Qd5! (winning for
White perhaps?) the finger of suspicion points at Black's 15th
move which may not deserve that exclamation mark.

Instead of 15...Be6 Black is usually given the sensible move 15...0-
0 e.g. 16 Ne7+ Kh8 17 Qd1 Ne4+ 18 Kb3 Qe6+ 19 Qd5 Nc5+. In
this line from the Handbuch, White's 17th is clearly weak but 17
Nxc8 Raxc8 is probably not much better. A critical line would
seem to be 17 Bf4 c6 (not 17...Be6 18 Qg2) when after 18 Nxc6
bxc6 19 Bxc6 Be6 even the exchange of queens should not save
White, e.g. 20 Qf3 Qxf3+ 21 Bxf3 Rad8 and White (despite two
extra pawns) will be hard pressed to save his king. Look at that
"pawn" on f4!

The other known possibility from the last diagram is 11 Nd4+ c6
when if 12 g3 Black plays 12...Qh3+ (not 12...Bh3+ as misprinted
in the final edition of my Chess Digest book on the Evans and Two
Knights) 13 Ke1 Qg2 when White has tactics with 14 Bxc6+ bxc6
15 Qxc6+ but they rebound after 15...Kf8 16 Rf1 Bxd4 17 Qxa8
Bf2+ (to this point, following the Handbuch) 18 Ke2 Nxg3+ 19
hxg3 Qxa8-+.

For much more detail on the known lines from the diagram, please
consult the excellent article by Mark Morss already referred to.

In Morss's own game, his opponent produced the genuine novelty
11 Be3!! (with the idea of meeting 11...Bxe3 by 12 g3) which is a
winning try for White, although it may be ultimately unsatisfactory.
(See Diagram)

The game continued 11...Ng3+!? 12 hxg3 Qxh1+ 13 Kf2 (13 Ke2?
Bg4+) 13...Bxe3+ 14 Kxe3 0-0 and Black eventually won but
White missed a drawing line at one point. Again, see the Morss
website for detailed analysis of the line, and a download of the
game.

I found that the computer program Fritz5, running overnight in
"correspondence analysis" mode had two suggestion to improve
Black's play in this game. Instead of 14...0-0 it favoured Black's
chances with 14...Qh6+! and it also showed a preference for 11...0-
0 over 11...Ng3+. 

Of course, this is all great fun but not of much practical use since 8
Nxc6 is so rarely played. However, maybe as Black you may meet
somebody who is following Emms, in which case you may be able
to take a point off them.

I think Emms does better when he turns his attention to the line (1
e4 e5 2 Nf3 Nc6 3 Bc4 Nf6 4 d4 exd4) 5 0-0 and considers both
5...Nxe4 and 5...Bc5 in some detail. (A strict "repertoire book"
would stick to one or the other but Emms takes his own paths and
recommends defending the Max Lange to the "adventurous
player").

I must say that I am very much on his side in this debate. On my
bookshelf I have a book called Winning With The Giuoco Piano
and Max Lange Attack by GM Soltis, but if the Max Lange is
winning for anyone (since many main lines lead to a draw) it is
Black who gets the point on the scoreboard, not White. Similarly,
English writer Chris Baker made the Max Lange one of the main
planks for White in his 1998 book Surprising Repertoire for White.
That alone almost made the book a non-starter for me, but it had to
be completely dismissed when I found After 1 e4 e5 2 Nf3 Nc6 3
Bc4 Nf6 4 d4 exd4 5 0-0 Bc5 (Emms also covers 5...Nxe4
adequately, I think,) 6 e5 d5 7 exf6 dxc4 8 Re1+ 

This is the venerable Max Lange Attack. Emms only analyses the
main line 8...Be6 but Black can instead play 8...Kf8 e.g. 9 Bg5
gxf6 10 Bh6+ Kg8 11 Nc3 Bf8 when White has a little-known
possibility, 12 Nxd4 (instead of Bxf8). (See Diagram)

Chris Baker (on page 9 of his 1998 book A Startling Opening
Repertoire says "12 Nxd4? Nxd4! 13 Qxd4 Qxd4 14 Re8 Qd6 15
Ne4 Bf5! 16 Nxd6 Rxe8 is winning for Black". However, a few
seconds with a program like Fritz will soon show you that White
mates in five by 15 Nd5!. A truly "startling" howler, which I
pointed out in my review at the time. I see that some other
reviewers praised the book without spotting this.

Instead of 13...Qxd4, Muller-Bayer, 1908, went 13...Bf5 14 Qf4
Bxc2 15 Rad1 Bd6 16 Nd5 Bxf4 17 Re8+ Qxe8 18 Nxf6 mate.

In fact it is Black's 12th move in Baker's variation, not White's, that
should have a question mark since 12...Bh6! leads to a queen swap
and roughly equal chances13 Nxc6 Qxd1 14 Ne7+ Kg7 15 Raxd1
Be6 16 Ned5 - M.Ford-J.King, Scottish open 1995. 

Instead of 8...Kf8, the normal move is 8...Be6 when White replies
9 Ng5. (See Diagram)

Baker and Soltis both also fails to mention the interesting
possibility for Black after 9 Ng5 of 9...Bf8!? which has been
successfully played by the current CC world champion Tonu Oim.
(Emms doesn't mention it either, but its omission in a repertoire
book for WHITE is far more serious.)

However, Black usually plays 9...Qd5. The lines this leads to are so
complex that one false step or miscalculation can lead to instant
reversal but I believe Emms is right that the general trend is in
Black's favour. After 10 Nc3 Qf5 11 Nce4 0-0-0 12 g4 Qe5 13
Nxe6 fxe6 14 fxg7 (not the only possibility) 14...Rhg8 15 Bh6 d3
16 c3 d2 17 Re2 Rd3 White's best chance is 18 Qf1. Emms gives a
very interesting correspondence game that I have not seen before to
show that 18 Nxc5 may be virtually a forced win for Black
18...Qxc5 19 Rxd2 Ne5 20 Rxd3 cxd3 21 Kg2 (He suggests 21
h3!? Qd5 22 Kh2 as a possible improvement.) 21...Qd5+ 22 Kg3
Nf7! 23 Qd2 Qd6+! 24 Kg2 e5 25 g5 Qg6! 

So far nothing surprising. Emms is wrong to say, however, that
White stands better after 25...Nd8!? 26 Rd1 Ne6 27 g6. In this
position, after 27...Rxg7!! 28 Bxg7 Nf4+ 29 Kg3 Black has at least
a draw.

However, he is right to say that 25...Qg6! is stronger. After 26 Kf1
Qe4 27 Kg1 Qg4+ 28 Kf1 Qf3 29 Kg1 Black has 29...e4! and this
wins, as in the game he cites (J.Moore-M.Read, corr England
1981). This winning idea was also discovered by French player
Laurent Tinture, the IECG game archivist, in a game he won more
recently.

Emms also deals with the 4 Ng5 d5 5 exd5 Na5 main line of the
Two Knights, where his advice seems trustworthy on the whole.
However, I did note that he cited one of my games and came to a
rather different opinion from the players. In fact very little actually
happens in this gameit is all "unheard melodies".

Tim Harding - Mike Read CC Olympiad preliminaries, 1992-93 1
e4 e5 2 Nf3 Nc6 3 Bc4 Nf6 4 Ng5 d5 5 exd5 Na5 6 Bb5+ c6 7
dxc6 bxc6 8 Be2 h6 9 Nf3 e4 10 Ne5 Bc5 

My opponent in this game was the same CC-IM Mike Read who
won the Max Lange game just mentioned. I expected his 10th
move, which he had played in some previously published games,
and I wanted to try out an idea I had at move 15.

11 c3 Qc7 12 f4 Bd6 13 d4 exd3 14 Qxd3 0-0 15 Nd2!?

This is a new move which gets queenside development going and
threatens to hold the Pawn by Ndf3. Read commented later, "An
interesting idea, and I think it was just as strong as the usual 15 O-
O."

15...Bxe5 16 fxe5 Qxe5 17 0-0

Emms comments: "White possesses the bishop pair but Black's
lead in development neutralises this". Neither Read nor I thought it
was so clear-cut. 

17...Re8!

The threat to e2 limits White's choice, ruling out Nf3.

18 Bd1! Bg4 19 Nf3 Qc7 20 Nd4 

Although he eliminates the two bishops, my development can now
catch up. If now 20...Bxd1 21 Bf4 is promising.

20...c5 - (See Diagram)

 The game was agreed drawn on Read's proposal but perhaps I
should have played on. He wrote to me''In the final position I 'felt'
uncomfortable - your QB was going to be a strong piece while my
QN was in limbo. Although my concrete analysis told me that I
was OK I just didn't like my position."

Maybe somebody else would like to test this line?