The Kibitzer by Tim Harding
     
The Eternal Appeal Of The Urusov Gambit

Last month's column dealt with the opening 1 e4 e5 2 Bc4 and I
mentioned that one option for White, after Black's normal reply
2...Nf6, is the central strike, 3 d4. I promised to come back to this,
so here we are.

I got onto the Internet, around New Year 1996, and discovered the
University of Pittsburgh Archive. One of the very first things I
discovered there was an article about the Urusov Gambit, which
caught my eye because this was one of the little-known gambits
that featured in my very first chess book, "Bishop's Opening" 25
years ago.

There were several Russian chessplayers called Urusov. The
gambit 1 e4 e5 2 Bc4 Nf6 3 d4 exd4 4 Nf3 is named after Prince
Sergey Semyonovich Urusov who was born on August 3, 1827 and
lived until November 20, 1897 (old-style calendar, I think). (See
Diagram)

Sergey Urusov, who became a Major-General in the Army, was the
elder brother of another strong player, Dmitry Urusov (1829-1903).
Interestingly, the Soviet Chess Encyclopaedia describes Sergey as
a friend of Tolstoy and an originator of new ideas in various
classical open games including the King's Gambit and the Scotch.
Unfortunately games of his with the gambit do not appear to have
survived.

Dmitry is described as a player of positional style - and a friend of
the writer Turgenev! (In a later generation, but not mentioned in
the Encyclopaedia, was N. Urusov who played at least one
correspondence game against Alekhine in the first decade of the
20th century.)

A number of players seemed interested in reviving this gambit, but
the file posted at Pittsburgh in those days was a textfile not easily
converted to PGN. Later, when I started my website, one of the
people behind that article, Max Burkett, contacted me but I was too
busy at the time to look further into the gambit. Now I see that a
zipfile of Urusov games is available to download from there in
Chessbase or PGN formats. 

It includes quite a few relatively recent games that show there is a
lot of interest in this gambit among amateurs now. However the
quality of the data is not that reliable, to judge from the following
miniature.

Burkett,M - Lawless,K Mechanics Institute Weekly, San Francis,
1977 1 e4 e5 2 Bc4 Nf6 3 d4 exd4 4 Nf3 d5 5 exd5 Bb4+ 6 Kf1 c5
7 a3 Ba5 8 b4 Bb6 9 bxc5 Bxc5 10 Nxd4 Nxd5 11 Nb3 1-0??

I don't understand this being given as a White win in the Pitt
database, since 11...Ne3+ wins at once for Black. 

The Urusov Gambit is actually an interesting challenge for analysts
and a good training ground for players wanting to improve their
understanding of tactics. My main analysis of the critical line will
come next month, but I want to give now one example of what it is
all about.

Let us look at one of the classic White successes with this gambit.

Schlechter v Neustadtl & Tietz, consultation game, Carlsbad 1901

1 e4 e5 2 Bc4 Nf6 3 d4 exd4 4 Nf3 Nxe4 5 Qxd4 Nf6 6 Nc3 Be7 7
Bg5 c6 

7...Nc6 is also an important line, as we shall see in the October
Kibitzer. 7...0-0 might seem attractive but the Russian analyst and
correspondence master remarks thar "Black determines
prematurely the position of his king" - Neishtadt. We shall see next
month that castling queenside can also be an important option in
this gambit.

8 0-0-0 d5 9 Rhe1 

9 Qh4 Be6 10 Bd3 Nbd7 11 Nd4 Nc5 12 Rhe1 is an alternative
move order but it doesn't give Black the opportunity of falling into
the trap seen in the present game.

9...0-0 

9...Be6 10 Bd3 (or 10 Qh4 Nbd7 transposing) 10...Nbd7 11 Qh4
Nc5 12 Nd4 is the other classical Urusov Gambit position for
analysis.

10 Qh4 (See Diagram)

White sets up a pin on the d-file and prepares kingside threats later. 

Now we have the central mystery of the Urusov Gambit. White is a
pawn done, apparently for next to nothing. He has no central
pawns, no structural advantage of any kind. If anything, the
d-pawn means that Black in the long run will have more space. 

Compare this with the King's Gambit (1 e4 e5 2 f4 exf4), where he
gets a majority of central pawns and can expect to regain the
f4-pawn at some stage, or even the Evans Gambit (1 e4 e5 2 Nf3
Nc6 3 Bc4 Bc5 4 b4 Bxb4 5 c3) where at least we see White
gaining time by attacking a displaced piece. 

So in what does White's compensation consist? It seems to be
strictly a matter of piece activity and open lines, so White must
play very actively and precisely to force a concession fine a short
time before Black can catch up on development. That is part of the
appeal of the Urusov Gambit. In this way, it rather resembles a
very modern gambit (The Belgrade Gambit, 1 e4 e5 2 Nf3 Nc6 3
Nc3 Nf6 4 d4 exd4 5 Nd5) where again White relies entirely on
time and tactics.

How Black can easily go wrong, we are now about to see.

10...h6? 

10...Be6 is correct, as we shall see next month. 

White now immediately exploited Black's slip by a fine
combination.

11 Bxd5!

With one blow, White regains his gambit pawn, takes out his
opponent's central bastion and opens the d-file. 

Black cannot accept the piece sacrifice because if 

a) 11...hxg5? 12 Bxf7+ Kxf7 (if 12...Rxf7 then 13 Rxd8 is check.)
13 Nxg5+ with a strong attack and no material deficit.

b) 11...cxd5 12 Nxd5! with decisive geometry.

11...Nbd7 12 Bc4 b5 13 Bd3 hxg5 

Else Black will have no material as consolation for White's very
strong attack.

14 Nxg5 Re8 15 Bh7+ Kf8 16 Bf5! Kg8 17 Nxf7! Kxf7 18 Be6+
Kg6 19 f4 Nh5 

The only move. Now, amazingly after his brilliant play, Schlechter
overlooked a forced mate in 5! He played 20 Qg4+? and took 49
moves to win. He could have produced a worthy finish by 20
Bf5+! Kf7 (20...Kxf5 21 g4+ Kxf4 22 Rf1+) 21 Qxh5+ g6
(21...Kf8 22 Be6) 22 Qh7+ Kf8 23 Qh8+ Kf7 24 Be6#.

(Note that the second player on the Black side was not Mieses, as
the Pitt database would have you believe. If Mieses had been
playing, Schlechter might not have had such an easy time.)

That is all about the main line until next month, as I need to
prepare those of you who may want to play the Urusov Gambit for
some early divergences that you could meet. Next month's column,
to be entitled "Is The Urusov Gambit Sound?", will study the
4...Nxe4 lines in detail. So if you have played interesting examples
please send them in during the next fortnight!

The main problem for players who want to try the Urusov Gambit
is that you must find someone willing to defend it, as Black can
circumvent the gambit in various ways.

Now let us look, in a more analytical way, at the introductory
moves to the Urusov Gambit, following 1 e4 e5 2 Bc4 Nf6 3 d4.
(See Diagram)

In this position, you can Black to play 3...exd4 although it is not
forced. The alternatives are 3....d5, 3...Nc6, 3...c6!? and 3...Nxe4.
These are not really germane to the main theme so let's get them
out of the way quickly

a) 3...d5 was suggested as playable by the late David Hooper in his
"A Complete Defence to 1 P-K4" but it looks like a luxury Black
should not be able to afford. Tartakower-von Scheve, Barmen
1905, continued 4 dxe5 Nxe4 5 Bxd5 Ng5 6 Bxg5 Qxg5 7 Nf3
Qh5 8 Nc3 Be7 9 h3 c6 10 Bb3 0-0 11 Qe2 Na6 12 Ne4 Nc5 13
Nxc5 Bxc5 14 0-0-0 b5 15 g4 Qg6 16 Nd2 Be6 17 f4 Bxb3 18
Nxb3 Be7 19 h4 Qe6 20 f5 Qc8 21 Qf3 c5 22 f6 Bd8 23 fxg7 Re8
24 Rhf1 Qe6 25 Nxc5 Qxa2 26 e6 f6 27 Qxa8 Qa1+ 28 Kd2 Qxb2
29 Qd5 Ba5+ 30 Ke3 Bb6 31 Kf3 Qc3+ 32 Nd3 Qxc2 33 Qd7 1-0.

b) If 3...Nc6, don't choose 4 Bxf7+?! as Sir George Thomas v
Milner-Barry, Hastings 1938, shows: 4... Kxf7 5 dxe5 Nxe5 6 f4
Nc6 7 e5 d5 8 exf6 Qxf6 9 Nf3 Bb4+ 10 c3 Re8+ 11 Kf1 Bd6 12
Ng5+ Kg6 13 Qd3+ Qf5 14 Qg3 Qg4 15 Qxg4 Bxg4 16 h3 Be2+
17 Kf2 Bc5+ 18 Kg3 Ne7 19 Kh2 Nf5 20 g4 Ne3 21 Na3 Bd6 22
Kg3 h6 23 Nf3 Re4 24 Kf2 Bxf3 25 Kxf3 Bxf4 26 Bxe3 Rxe3+ 27
Kf2 (27 Kxf4 Rae8) 27...Rae8 28 Rad1 Re2+ 29 Kf3 Bg5 30 Rd3
Bh4 31 Rf1 0-1

However 4 dxe5 should give White the edge, e.g. 4...Nxe5 (not
4...Nxe4? 5 Bxf7+ Kxf7 6 Qd5+) 5 Be2 where the bishop covers
both wings and later the black knight, which has no central anchor
point, can be pushed back by f4 or exchanged by Nf3, with gain of
time in either case. An example from the Pitt database is
Schaefer-Schulenburg, 1994, which went 4 dxe5 dxc4 5 Qxd8+
Kxd8 6 exf6 gxf6 7 Nc3 Bd6 8 Nge2 Rg8 9 Ng3 Bg4 10 f3 Be6 11
Be3 c6 12 0-0-0 Kc7 13 Nce2 and White's safer king and sounder
pawn structure should carry more weight than Black's bishop pair.
White eventually won after some complications.

White has also tried 4 d5 which might suit players who like the
kind of blocked centres that can also arise from the Hungarian
Defence.

c) 3...c6 was punished in the game Marshall-Forsberg, Dimock
Theme Tournament, New York, 1924 dxe5 Qa5+ 5 Nc3 Nxe4 6
Bxf7+ Kxf7 7 Qf3+ Nf6 8 exf6 Qe5+ 9 Ne4 Bb4+ 10 Kf1 Re8 11
fxg7+ Kxg7 12 Qg4+ Kh8 13 Bg5 Be7 14 Nf3 Qb5+ 15 Kg1 d5 16
Qh5 Bd7 17 Nf6 1-0.

d) The most likely of the alternatives from the diagram is 3...Nxe4
4 dxe5 when 4...Nc5 (Jaenisch) is clearly better for White.
Neishtadt-Gipslis, USSR 1955, went on 5 Nf3 Be7 6 Bf4 Nc6 7
Nc3 Ne6 8 Bg3 0-0 9 Qe2 f5 10 0-0-0 Qe8 11 Nd5 Kh8 12 Nf4 a6
13 h4 Na5 14 Bxe6 dxe6 15 Ng5 Bc5 16 Rd8 Qxd8 17 Qh5 h6 18
Qg6 hxg5 19 hxg5+ Kg8 20 Qh5 Nc4 21 g6 Qd2+ 22 Kb1 Na3+
23 bxa3 1-0. Later Gipslis became a grandmaster.

Also after 3..Nxe4 4 dxe5 a line from Staunton went 4...Bc5? 5
Bxf7+ Kf8 6 Qf3. Instead of 4...Nc5 and 4...Bc5, Black can create
a crude threat by 4...Qh4 but after 5 Qf3 Ng5
(Dorfman-Zilberstein, USSR 1974) Larsen recommended 6 Qf4
Qxf4 7 Bxf4 Ne6 8 Bg3 when White has a development lead and
the e-pawn cramps his opponent.

So, to return to the line 1 e4 e5 2 Bc4 Nf6 3 d4, Black normally
plays 3...exd4. Then the old line for White, given by Ponziani in
the 18th century, was 4 e5 but I am not interested in that; it is
clearly time-wasting and anti-positional. Simply 4...d5 is a good
reply.

The move for White to play is Sergey Urusov's true innovation, 4
Nf3 (See Diagram).

This position can also arise via the Petroff Defense, 1 e4 e5 2 Nf3
Nf6 3 d4 exd4 4 Bc4 (instead of 4 e5), but in practice it is rare to
get it via that route because Black almost invariably plays the
fashionable 3...Nxe4 in that case.

Now 4...Nxe4, in the diagram position, 5 Qxd4! is the Urusov
Gambit Accepted. 

This is Black's second and last chance to decline it.

It is important to realise that you have to be ready to meet the Two
Knights Defence, because Black may well (and often does) simply
play 4...Nc6 here. You can reply 5 0-0 or 5 e5, according to taste,
or maybe the less favoured 5 Ng5, but anyway that line is beyond
the scope of this article. However, if you know your opponent
normally meets 1 e4 e5 2 Nf3 Nc6 3 Bc4 by 3...Bc5 (or 3...Be7 or
3...d6) rather than 3...Nf6, then he or she is a good bet for the
Urusov.

Black can also play 4...Bb4+, 4...Bc5 or 4...d5. (I have even see
4...c5?! e.g. 5 Qe2 d6 6 e5 dxe5 7 Nxe5 Be6 8 Nxf7 Qe7 9 Qxe6 1-0 
Akos- Csaszar, corr 1968).

The first three moves all deserve more attention. 

After 4...Bb4+ Euwe recommended 5 c3 dxc3 6 bxc3 intending 7
e5 with good chances; Larsen added the note 6...Ba5 7 e5 Qe7 8
Ba3. However, a Burkett-Malmstrom IECG game saw 6....d5! So I
suspect the correct answer is the real gambit style 6 0-0! which has
been played the masters Horowitz and Acers. See the Pitt database
for examples.

As for 4...Bc5, this brings us close to Italian game territory. White
usually plays 5 0-0. In that case, Black can transpose to the Max
Lange Attack by 5...Nc6 or play other moves such as 5...d6 (which
might be met by 6 b4!?). However, there is an alternative in 5 e5
(relatively little investigated) when the critical line goes 5...d5
(White has been winning regularly against knight moves.) 6 exf6
dxc4 7 Qe2+ Be6 8 fxg7 Rg8 9 Bg5 Qd5 (9...Be7 also needs
analysis) with a position rather like a Max Lange except that Black
has not played ...Nc6. (See Diagram)

Now there was a 1988 correspondence game Vavrovski-Stieg
which went 10 Nbd2 from the diagram; Black won. White should
analyse an old Euwe suggestion of 10 Nc3!? intending 0-0-0. If
Black accepts the sacrifice by 10...dxc3 11 Rd1 cxb2 (11...Bxf2+
12 Kf1!) then 12 0-0 b1Q 13 Rxd5 seems to be winning for White.
So 5 e5 is definitely worth following up.

Finally, there is 4...d5 of which I have personal experience. This
gives good open tactical play where White should hold the
initiative although it must be possible for Black to defend better
than in the following game.

Tim Harding - J. Boyce Woolacombe, 1973 1 e4 e5 2 Nf3 Nf6 3
d4 exd4 4 Bc4 d5 5 exd5 Nxd5? 

I also once met 5...c5?! 6 0-0 Nxd5 (6...Bg4 7 c3!) 7 c3 Nc6
(7...Nb6!?) 8 Qb3 Na5 9 Bb5+ with some advantage to White
(Harding-Risdon, corr 1987).

5...Bb4+ is the critical line6 Bd2!? was once suggested to me by
Max Zavanelli, and has been successfully tried. However, 6 c3 is
probably strong6...Qe7+ 7 Be2 (7 Kf1 is about equal.) 7...dxc3 8
bxc3 Bc5 9 0-0 0-0 and now 10 Bg5! (Instead of 10 c4= as given in
ECO) 10...h6 11 Bh4 Rd8 12 Re1 g5 13 Bc4 Be6 14 Nxg5! Ng4
15 Nf3 Qd6 16 h3! Rd7 17 hxg4 f6 18 Qc1 1 0 P.Grave-P.Zaiser,
West Germany - USSR corr match 1965-67.

6 0-0 Nc6?! 

6...Be7 is probably better.

7 Re1+ Be7 8 Nxd4 Nxd4?! 9 Qxd4 Be6 10 Qxg7 Bf6 11 Qh6
Rg8 12 Nc3!

After this the game is fun only for White!

12...Nxc3 13 bxc3

13 Bxe6? Ne2+!

13...Bxc3? 

13...Rg6 14 Qxh7 also wins for White.

14 Rxe6+! fxe6 

If 14...Kd7 15 Qh3! wins.

15 Qxe6+ Qe7 16 Qxg8+ Kd7 17 Qd5+ Ke8 18 Bg5 c6 19 Qg8+
Qf8 20 Qe6+ 1-0

Copyright 1998 Tim Harding. All Rights Reserved.

